AI Data Trainer/annotator

AI Data Trainer/annotator — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

  • Physical access

    Physical access

    Physical access is a term in computer security that refers to the ability of people to physically gain access to a computer system. According to Gregory White, "Given physical access to an office, the knowledgeable attacker will quickly be able to find the information needed to gain access to the organization's computer systems and network." == Attacks and countermeasures == === Attacks === Physical access opens up a variety of avenues for hacking. Michael Meyers notes that "the best network software security measures can be rendered useless if you fail to physically protect your systems," since an intruder could simply walk off with a server and crack the password at his leisure. Physical access also allows hardware keyloggers to be installed. An intruder may be able to boot from a CD or other external media and then read unencrypted data on the hard drive. They may also exploit a lack of access control in the boot loader; for instance, pressing F8 while certain versions of Microsoft Windows are booting, specifying 'init=/bin/sh' as a boot parameter to Linux (usually done by editing the command line in GRUB), etc. One could also use a rogue device to access a poorly secured wireless network; if the signal were sufficiently strong, one might not even need to breach the perimeter. === Countermeasures === IT security standards in the United States typically call for physical access to be limited by locked server rooms, sign-in sheets, etc. Physical access systems and IT security systems have historically been administered by separate departments of organizations, but are increasingly being seen as having interdependent functions needing a single, converged security policy. An IT department could, for instance, check security log entries for suspicious logons occurring after business hours, and then use keycard swipe records from a building access control system to narrow down the list of suspects to those who were in the building at that time. Surveillance cameras might also be used to deter or detect unauthorized access.

    Read more →
  • Rumelhart Prize

    Rumelhart Prize

    The David E. Rumelhart Prize for Contributions to the Theoretical Foundations of Human Cognition was founded in 2001 in honor of the cognitive scientist David Rumelhart to introduce the equivalent of a Nobel Prize for cognitive science. It is awarded annually to "an individual or collaborative team making a significant contemporary contribution to the theoretical foundations of human cognition". The annual award is presented at the Cognitive Science Society meeting, where the recipient gives a lecture and receives a check for $100,000. At the conclusion of the ceremony, the next year's award winner is announced. The award is funded by the Robert J. Glushko and Pamela Samuelson Foundation. The Rumelhart Prize committee is independent of the Cognitive Science Society. However, the society provides a large and interested audience for the awards. == Selection Committee == As of 2022, the selection committee for the prize consisted of: Richard Cooper (chair) Dedre Gentner Robert J. Glushko Tania Lombrozo Steven T. Piantadosi Jesse Snedeker == Recipients ==

    Read more →
  • Document AI

    Document AI

    Document AI, also known as Document Intelligence, refers to a field of technology that employs machine learning (ML) techniques, such as natural language processing (NLP). These techniques are used to develop computer models capable of analyzing documents in a manner akin to human review. Through NLP, computer systems are able to understand relationships and contextual nuances in document contents, which facilitates the extraction of information and insights. Additionally, this technology enables the categorization and organization of the documents themselves. The applications of Document AI extend to processing and parsing a variety of semi-structured documents, such as forms, tables, receipts, invoices, tax forms, contracts, loan agreements, and financial reports. == Key features == Machine learning is utilized in Document AI to extract information from both printed and digital documents. This technology recognizes images, text, and characters in various languages, aiding in the extraction of insights from unstructured documents. The use of this technology can improve the speed and quality of decision-making in document analysis. Additionally, the automation of data extraction and validation can contribute to increased efficiency in document analysis processes. Since the early 2020s, the integration of large language models has extended Document AI beyond extraction toward generative tasks, including the automated drafting of forms, contracts, and document summaries. == Example == A business letter contains information in the form of text, as well as other types of information, such as the position of the text. For instance, a typical letter contains two addresses before the body of the text. The address at the very top (sometimes aligned to the right) is the sender address. This is normally followed by the date of the letter, with the place of writing. After this, the receiver address is listed. The distinction between the sender address and the receiver address is conveyed solely by the position of the address on the page, i.e. there is no textual indication like Sender: in front of the addresses. == Data dimensions and ML architecture == Data is typically distinguished into spatial data and time-series data, the former includes things like images, maps and graphs, while the latter includes signals such as stock prices or voice recordings. Document AI combines text data, which has a time dimension, with other types of data, such as the position of an address in a business letter, which is spatial. Historically in machine learning spatial data was analyzed using a convolutional neural network, and temporal data using a recurrent neural network. With the advent of dimension-type agnostic transformer architecture, these two different types of dimension can be more easily combined, Document AI is an example of this. == Benchmarks == Several public datasets are used to evaluate Document AI systems. FUNSD (Form Understanding in Noisy Scanned Documents) contains 199 annotated forms with token- and block-level labels for form understanding tasks. CORD (Consolidated Receipt Dataset) supports key information extraction from receipts. DocVQA contains approximately 50,000 questions over 12,000 document images for layout-aware visual question answering. == Common uses == Document AI systems are used to automate document processing and information extraction in business and financial workflows, including invoice and receipt processing, data entry automation, anomaly detection, mortgage processing, loan portfolio monitoring, credit risk management, and fraud detection such as counterfeit currency and fraudulent checks. They are also applied in regulatory compliance and contract analysis, including assessing changes in legal and regulatory documents. In real estate, Document AI supports document classification and structured information extraction for standardized processing and analytics. With the adoption of generative AI, Document AI systems can also generate and pre-fill structured documents such as contracts or business forms from natural language prompts.

    Read more →
  • AI art

    AI art

    Artificial intelligence visual art, or AI art, is visual artwork generated or enhanced through the implementation of artificial intelligence (AI) programs, most commonly using text-to-image models. The process of automated art-making has existed since antiquity. The field of artificial intelligence was founded in the 1950s, and artists began to create art with artificial intelligence shortly after the discipline's founding. A select number of these creations have been showcased in museums and have been recognized with awards. Throughout its history, AI has raised many philosophical questions related to the human mind, artificial beings, and the nature of art in human–AI collaboration. During the AI boom of the 2020s, text-to-image models such as Midjourney, DALL-E and Stable Diffusion became widely available to the public, allowing users to quickly generate imagery with little effort. Commentary about AI art in the 2020s has often focused on issues related to copyright, deception, defamation, and its impact on more traditional artists, including technological unemployment. In August 2023, the US Supreme Court ruled that AI art is ineligible for copyright due to failure to meet human authorship. In March 2026, it declined to hear a case over whether AI-generated art can be subject to copyright. == History == === Early history === Automated art dates back at least to the automata of ancient Greek civilization, when inventors such as Daedalus and Hero of Alexandria were described as designing machines capable of writing text, generating sounds, and playing music. Creative automatons have flourished throughout history, such as Maillardet's automaton, created around 1800 and capable of creating multiple drawings and poems. Also in the 19th century, Ada Lovelace, wrote that "computing operations" could potentially be used to generate music and poems. In 1950, Alan Turing's paper "Computing Machinery and Intelligence" focused on whether machines can mimic human behavior convincingly. Shortly after, the academic discipline of artificial intelligence was founded at a research workshop at Dartmouth College in 1956. Since its founding, AI researchers have explored philosophical questions about the nature of the human mind and the consequences of creating artificial beings with human-like intelligence; these issues have previously been explored by myth, fiction, and philosophy since antiquity. === Artistic history === Since the founding of AI in the 1950s, artists have used artificial intelligence to create artistic works. These works were sometimes referred to as algorithmic art, computer art, digital art, or new media art. One of the first significant AI art systems is AARON, developed by Harold Cohen beginning in the late 1960s at the University of California at San Diego. AARON uses a symbolic rule-based approach to generate technical images in the era of GOFAI programming, and it was developed by Cohen with the goal of being able to code the act of drawing. AARON was exhibited in 1972 at the Los Angeles County Museum of Art. From 1973 to 1975, Cohen refined AARON during a residency at the Artificial Intelligence Laboratory at Stanford University. In 2024, the Whitney Museum of American Art exhibited AI art from throughout Cohen's career, including re-created versions of his early robotic drawing machines. Karl Sims has exhibited art created with artificial life since the 1980s. He received an M.S. in computer graphics from the MIT Media Lab in 1987 and was artist-in-residence from 1990 to 1996 at the supercomputer manufacturer and artificial intelligence company Thinking Machines. In both 1991 and 1992, Sims won the Golden Nica award at Prix Ars Electronica for his videos using artificial evolution. In 1997, Sims created the interactive artificial evolution installation Galápagos for the NTT InterCommunication Center in Tokyo. Sims received an Emmy Award in 2019 for outstanding achievement in engineering development. In 1999, Scott Draves and a team of several engineers created and released Electric Sheep as a free software screensaver. Electric Sheep is a volunteer computing project for animating and evolving fractal flames, which are distributed to networked computers that display them as a screensaver. The screensaver used AI to create an infinite animation by learning from its audience. In 2001, Draves won the Fundacion Telefónica Life 4.0 prize for Electric Sheep. In 2014, Stephanie Dinkins began working on Conversations with Bina48. For the series, Dinkins recorded her conversations with BINA48, a social robot that resembles a middle-aged black woman. In 2019, Dinkins won the Creative Capital award for her creation of an evolving artificial intelligence based on the "interests and culture(s) of people of color." In 2015, Sougwen Chung began Mimicry (Drawing Operations Unit: Generation 1), an ongoing collaboration between the artist and a robotic arm. In 2019, Chung won the Lumen Prize for her continued performances with a robotic arm that uses AI to attempt to draw in a manner similar to Chung. In 2018, an auction sale of artificial intelligence art was held at Christie's in New York where the AI artwork Edmond de Belamy sold for US$432,500, which was almost 45 times higher than its estimate of US$7,000–10,000. The artwork was created by Obvious, a Paris-based collective. In 2024, Japanese film generAIdoscope was released. The film was co-directed by Hirotaka Adachi, Takeshi Sone, and Hiroki Yamaguchi. All video, audio, and music in the film were created with artificial intelligence. In 2025, the Japanese anime television series Twins Hinahima was released. The anime was produced and animated with AI assistance during the process of cutting and conversion of photographs into anime illustrations and later retouched by art staff. Most of the remaining parts such as characters and logos were hand-drawn with various software. === Technical history === Deep learning, characterized by its multi-layer structure that attempts to mimic the human brain, first came about in the 2010s, causing a significant shift in the world of AI art. During the deep learning era, there are mainly these types of designs for generative art: autoregressive models, diffusion models, GANs, normalizing flows. In 2014, Ian Goodfellow and colleagues at Université de Montréal developed the generative adversarial network (GAN), a type of deep neural network capable of learning to mimic the statistical distribution of input data such as images. The GAN uses a "generator" to create new images and a "discriminator" to decide which created images are considered successful. Unlike previous algorithmic art that followed hand-coded rules, generative adversarial networks could learn a specific aesthetic by analyzing a dataset of example images. In 2015, a team at Google released DeepDream, a program that uses a convolutional neural network to find and enhance patterns in images via algorithmic pareidolia. The process creates deliberately over-processed images with a dream-like appearance reminiscent of a psychedelic experience. Later, in 2017, a conditional GAN learned to generate 1000 image classes of ImageNet, a large visual database designed for use in visual object recognition software research. By conditioning the GAN on both random noise and a specific class label, this approach enhanced the quality of image synthesis for class-conditional models. Autoregressive models were used for image generation, such as PixelRNN (2016), which autoregressively generates one pixel after another with a recurrent neural network. Immediately after the Transformer architecture was proposed in Attention Is All You Need (2018), it was used for autoregressive generation of images, but without text conditioning. The website Artbreeder, launched in 2018, uses the models StyleGAN and BigGAN to allow users to generate and modify images such as faces, landscapes, and paintings. In the 2020s, text-to-image models, which generate images based on prompts, became widely used, marking yet another shift in the creation of AI-generated artworks. In 2021, using the influential large language generative pre-trained transformer models that are used in GPT-2 and GPT-3, OpenAI released a series of images created with the text-to-image AI model DALL-E 1. It is an autoregressive generative model with essentially the same architecture as GPT-3. Along with this, later in 2021, EleutherAI released the open source VQGAN-CLIP based on OpenAI's CLIP model. Diffusion models, generative models used to create synthetic data based on existing data, were first proposed in 2015, but they only became better than GANs in early 2021. Latent diffusion model was published in December 2021 and became the basis for the later Stable Diffusion (August 2022), developed through a collaboration between Stability AI, CompVis Group at LMU Munich, and Runway. In 2022, Midjourney was released, followed by Google Brain's Imagen and Pa

    Read more →
  • C3D Toolkit

    C3D Toolkit

    C3D Toolkit is a proprietary cross-platform geometric modeling kit software developed by Russian C3D Labs (previously part of ASCON Group). It's written in C++ . It can be licensed by other companies for use in their 3D computer graphics software products. The most widely known software in which C3D Toolkit is typically used are computer aided design (CAD), computer-aided manufacturing (CAM), and computer-aided engineering (CAE) systems. C3D Toolkit provides routines for 3D modeling, 3D constraint solving, polygonal mesh-to-B-rep conversion, 3D visualization, and 3D file conversions etc. == History == Nikolai Golovanov is a graduate of the Mechanical Engineering department of Bauman Moscow State Technical University as a designer of space launch vehicles. Upon his graduation, he began with the Kolomna Engineering Design bureau, which at the time employed the future founders of ASCON, Alexander Golikov and Tatiana Yankina. While at the bureau, Dr Golovanov developed software for analyzing the strength and stability of shell structures. In 1989, Alexander Golikov and Tatiana Yankina left Kolomna to start up ASCON as a private company. Although they began with just an electronic drawing board, even then they were already conceiving the idea of three-dimensional parametric modeling. This radical concept eventually changed flat drawings into three-dimensional models. The ASCON founders shared their ideas with Nikolai Golovanov, and in 1996 he moved to take up his current position with ASCON. As of 2012 he was involved in developing algorithms for C3D Toolkit. In 2012 the earliest version of the C3D Modeller kernel was extracted from KOMPAS-3D CAD. It was later adopted to a range of different platforms and advertised as a separate product. == Overview == It incorporates five modules: C3D Modeler constructs geometric models, generates flat projections of models, performs triangulations, calculates the inertial characteristics of models, and determines whether collisions occur between the elements of models; C3D Modeler for ODA enables advanced 3D modeling operations through the ODA's standard "OdDb3DSolid" API from the Open Design Alliance; C3D Solver makes connections between the elements of geometric models, and considers the geometric constraints of models being edited; C3D B-Shaper converts polygonal models to boundary representation (B-rep) bodies; C3D Vision controls the quality of rendering for 3D models using mathematical apparatus and software, and the workstation hardware; C3D Converter reads and writes geometric models in a variety of standard exchange formats. == Features == == Development == == Applications == Since 2013 - the date the company started issuing a license for the toolkit -, several companies have adopted C3D software components for their products, users include: Recently, C3D Modeler has been adapted to ODA Platform. In April 2017, C3D Viewer was launched for end users. The application allows to read 3D models in common formats and write it to the C3D file format. Free version is available.

    Read more →
  • Fragile Dreams: Farewell Ruins of the Moon

    Fragile Dreams: Farewell Ruins of the Moon

    Fragile Dreams: Farewell Ruins of the Moon (フラジール ~さよなら月の廃墟~, Furajīru: Sayonara Tsuki no Haikyo; known in Japan as Fragile) is an action role-playing game for the Wii developed by Namco Bandai Games in co-operation with Tri-Crescendo. The game was released by Namco Bandai Games in Japan on January 22, 2009. It was later published by Xseed Games in North America on March 16, 2010, and in Europe by Rising Star Games on March 19, 2010, followed by its release in Australia on April 1, 2010. == Gameplay == In Fragile Dreams, the player character, Seto, must traverse the ruins of Tokyo and the surrounding areas, fighting off ghosts that lurk within these ruins. The game's heads-up display includes a mini-map and HP gauge for Seto's location and health, respectively. Seto will fall unconscious if his HP reaches zero, resulting in a game over. The player controls Seto from a third-person perspective with the Wii Remote and Nunchuk. Seto can use his flashlight (controlled by the Wii Remote pointer) to illuminate his surroundings or solve puzzles and interact with the environment. When searching for certain objectives or hidden enemies, pointing Seto's light in their direction picks up and plays their sounds through the Wii Remote's mini speaker. The Wii Nunchuk, meanwhile, directly controls Seto's movement: aside of basic movement, he can crouch to hide and crawl through small spaces. Seto will often come across damaged floors, which require slow movement (and for heavily damaged floors, crouching) to cross without falling through. As Seto, the player can use weapons found throughout the world to fight off ghosts, ranging from slingshots and golf clubs to crossbows and katanas. Each weapon can only take a certain amount of use: once a weapon reaches its limit, it will break after battle. The player can also find other usable and collectable items in the field, marked with fireflies. The player can only save their game by resting at small fire pits scattered throughout the world: used fire pits are marked with a bonfire. The player can also examine and identify Mystery Items, organize their inventory, as well as after encountering the Merchant, buy and sell items. As stated by the producer of the game, Kentarō Kawashima, Fragile Dreams is not strictly a survival horror: rather, its story focuses on human drama. In Fragile Dreams, aside of the main story, the player can find and examine objects and graffiti throughout the world. Objects called memory items (ranging from origami and stones to cell phones and books) hold the memories of their former owners (only accessible at bonfires), while the graffiti contains messages only seen by pointing at them in first-person. By examining these messages, the player can piece together hints to the game's backstory. == Story == === Setting and characters === Fragile Dreams is set in a post-apocalyptic version of Earth in the near-future. Almost all the world's population has vanished, leaving the surviving buildings and structures abandoned. The game is set in and near the ruins of Tokyo, Japan, where the event that nearly wiped out humanity may have originated. The protagonist, Seto, is a 15-year-old boy who searches the world for other living humans. He encounters Ren, a silver-haired girl who often leaves behind large, cryptic drawings. Other characters include: Sai, the ghost of a young woman; Crow, a mischievous and straightforward amnesiac boy; Personal Frame (P.F.), a portable computer who loves having conversations more than anything else; Chiyo, the ghost of a little girl; and the Merchant, a mysterious yet merry man who trades various goods. The game's host of enemies mainly consist of ghosts, but also include humanoid robots and security proxies. The main antagonist, Shin, is the AI of a scientist who considers speech to be an inferior means of communication. Various memory items include a greater set of characters, each giving hints to the game's backstory. === Plot === At the end of Seto's fifteenth summer, his grandfather dies. Seto buries him in front of their home, an old observatory, and that from then on he became "truly alone". At night, he searches for anything the old man had left for him and discovers a letter, along with a strange blue stone in a locket. Suddenly, a mask-like ghost appears and attacks Seto. After driving the creature off, Seto reads the old man's letter, who tells him to "reach a tall red tower" east of the observatory, where he might find other survivors. After departing for the tower, Seto reaches an old subway entrance in the Azabudai district and finds Ren sitting on a collapsed pillar, singing to the stars. He accidentally startles her and the frightened Ren flees into the subway station: getting over the shock of meeting another person, Seto follows her. While searching the station, he discovers a Personal Frame, who guides him towards Ren. Unfortunately, just as they reach the exit, P.F.'s battery dies out: Seto buries the device, keeping a screw from it in his locket. From the underground, Seto finds himself at an abandoned amusement park and encounters Crow, who steals Seto's locket. After a long chase across the park and another encounter with the masked ghost, Crow returns Seto's locket and directs him to a hotel nearby, where he saw a girl who might know something about Ren. Crow also gives Seto his skull ring to keep in his locket and kisses him. At the hotel, Seto encounters Sai and fights the masked ghost again. After laying to rest the spirit of an old woman named Chiyo, the two discover Ren's drawings by a sewer. Returning to the underground, Seto and Sai find themselves at a hydropower dam. While searching for Ren, Seto discovers that Crow is actually a robot, but his battery begins to fail and Seto mourns for him as he "die[s]". Finally, they encounter Ren in a cell: although glad to see him again, Ren runs off after Shin calls. Sai explains to Seto that most of humanity died because of a "human empathy expansion project" called Glass Cage. The project was meant to make human thoughts transparent, meaning that no one would need words to communicate. However, after Glass Cage activated, people who went to sleep never woke up again. Sai reveals that she was Glass Cage's first catalyst: this time, Shin intends to use Ren as the catalyst. After exiting the dam, a demolition crane attempts to destroy it. Hearing both Shin's and the masked ghost's voices from the crane — saying, "Any threat to the project must be eliminated." — the player realizes both are manifestations of Glass Cage. After Seto destroys the crane, Sai leads him to the facility where Ren was taken to. Entering the laboratory, Seto and Sai are confronted by Shin, who coldly dismisses Sai's attempts at reasoning with him and is adamant about proceeding with his plans. As they traverse the laboratory, they overhear a voice announcing "Glass Cage Launch Preparations Complete", strengthening their resolve to save Ren. Making it into the room where Ren is being held, Shin tells them of his intention to use Glass Cage to "obliterate corporeal beings". After Seto defeats him, Shin disappears and Seto releases Ren from the device holding her. Their reunion is cut short as Sai tells them that the backup system has "finished copying her psyche to the AI", allowing Glass Cage to proceed. Ren reveals Shin has escaped to the top of the Tokyo Tower and Seto asks Ren to wait at the base of the tower and for Sai to accompany her. On his way up the tower, Seto hears the voices of P.F., Chiyo and Crow wishing him luck. He confronts and defeats Shin a second time, who reveals his motivations: he had secretly used himself as the first test subject of the human empathy expansion project and gained the ability to hear the thoughts of those around him. Despite his initial belief in the project as a way for humans to empathize with one another, all he heard around him was "jealousy and contempt" and he soon grew disillusioned with the world as even his parents turned against him. Believing no person loved him, Shin wants to put an end to humanity. His words meet with a vehement response from Sai, as she tells him that she loves him, having developed those feelings while she was the catalyst and all she ever wanted was to be part of his life. Hearing this, Shin finds peace, tossing the AI mainframe away so Glass Cage can never be reactivated and vanishes together with Sai, hand-in-hand, after thanking Seto. Descending from the tower, Seto finally learns Ren's name and they resolve to look for other survivors together. == Development == Fragile Dreams was developed by the team at Namco Bandai Games. Director and producer Kentarō Kawashima came up with the concept for the game in 2003, before the Wii console was revealed. When the Wii was unveiled, it became the obvious choice as the game's platform as the Wii remote could be used to control the flashlight. Kawashima wrote the main scenario for the title, w

    Read more →
  • Polyworld

    Polyworld

    Polyworld is a cross-platform (Linux, Mac OS X) program written by Larry Yaeger to evolve Artificial Intelligence through natural selection and evolutionary algorithms. It uses the Qt graphics toolkit and OpenGL to display a graphical environment in which a population of trapezoid agents search for food, mate, have offspring, and prey on each other. The population is typically only in the hundreds, as each individual is rather complex and the environment consumes considerable computer resources. The graphical environment is necessary since the individuals actually move around the 2-D plane and must be able to "see." Since some basic abilities, like eating carcasses or randomly generated food, seeing other individuals, mating or fighting with them, etc., are possible, a number of interesting behaviours have been observed to spontaneously arise after prolonged evolution, such as cannibalism, predators and prey, and mimicry. Each individual makes decisions based on a neural net using Hebbian learning; the neural net is derived from each individual's genome. The genome does not merely specify the wiring of the neural nets, but also determines their size, speed, color, mutation rate and a number of other factors. The genome is randomly mutated at a set probability, which are also changed in descendant organisms.

    Read more →
  • Polyworld

    Polyworld

    Polyworld is a cross-platform (Linux, Mac OS X) program written by Larry Yaeger to evolve Artificial Intelligence through natural selection and evolutionary algorithms. It uses the Qt graphics toolkit and OpenGL to display a graphical environment in which a population of trapezoid agents search for food, mate, have offspring, and prey on each other. The population is typically only in the hundreds, as each individual is rather complex and the environment consumes considerable computer resources. The graphical environment is necessary since the individuals actually move around the 2-D plane and must be able to "see." Since some basic abilities, like eating carcasses or randomly generated food, seeing other individuals, mating or fighting with them, etc., are possible, a number of interesting behaviours have been observed to spontaneously arise after prolonged evolution, such as cannibalism, predators and prey, and mimicry. Each individual makes decisions based on a neural net using Hebbian learning; the neural net is derived from each individual's genome. The genome does not merely specify the wiring of the neural nets, but also determines their size, speed, color, mutation rate and a number of other factors. The genome is randomly mutated at a set probability, which are also changed in descendant organisms.

    Read more →
  • Desktop Window Manager

    Desktop Window Manager

    Desktop Window Manager (DWM, previously Desktop Compositing Engine or DCE in builds of pre-reset Windows Longhorn) is the compositing window manager in Microsoft Windows since Windows Vista that enables the use of hardware acceleration to render the graphical user interface of Windows. It was originally created to enable portions of the new "Windows Aero" user experience, which allowed for effects such as transparency, 3D window switching and more. It is also included with Windows Server 2008, but requires the "Desktop Experience" feature and compatible graphics drivers to be installed. == Architecture == The Desktop Window Manager is a compositing window manager, meaning that each program has a buffer that it writes data to; DWM then composites each program's buffer into a final image. By comparison, the stacking window manager in Windows XP and earlier (and also Windows Vista and Windows 7 with Windows Aero disabled) comprises a single display buffer to which all programs write. DWM works in different ways depending on the operating system (Windows 7 or Windows Vista) and on the version of the graphics drivers it uses (WDDM 1.0 or 1.1). Under Windows 7 and with WDDM 1.1 drivers, DWM only writes the program's buffer to the video RAM, even if it is a graphics device interface (GDI) program. This is because Windows 7 supports (limited) hardware acceleration for GDI and in doing so does not need to keep a copy of the buffer in system RAM so that the CPU can write to it. Because the compositor has access to the graphics of all applications, it easily allows visual effects that string together visuals from multiple applications, such as transparency. DWM uses DirectX to perform the function of compositing and rendering in the GPU, freeing the CPU of the task of managing the rendering from the off-screen buffers to the display. However, it does not affect applications painting to the off-screen buffers – depending on the technologies used for that, this might still be CPU-bound. DWM-agnostic rendering techniques like GDI are redirected to the buffers by rendering the user interface (UI) as bitmaps. DWM-aware rendering technologies like WPF directly make the internal data structures available in a DWM-compatible format. The window contents in the buffers are then converted to DirectX textures. The desktop itself is a full-screen Direct3D surface, with windows being represented as a mesh consisting of two adjacent (and mutually inverted) triangles, which are transformed to represent a 2D rectangle. The texture, representing the UI chrome, is then mapped onto these rectangles. Window transitions are implemented as transformations of the meshes, using shader programs. With Windows Vista, the transitions are limited to the set of built-in shaders that implement the transformations. Greg Schechter, a developer at Microsoft has suggested that this might be opened up for developers and users to plug in their own effects in a future release. DWM only maps the primary desktop object as a 3D surface; other desktop objects, including virtual desktops as well as the secure desktop used by User Account Control are not. Because all applications render to an off-screen buffer, they can be read off the buffer embedded in other applications as well. Since the off-screen buffer is constantly updated by the application, the embedded rendering will be a dynamic representation of the application window and not a static rendering. This is how the live thumbnail previews and Windows Flip work in Windows Vista and Windows 7. DWM exposes a public API that allows applications to access these thumbnail representations. The size of the thumbnail is not fixed; applications can request the thumbnails at any size - smaller than the original window, at the same size or even larger - and DWM will scale them properly before returning. Aero Flip does not use the public thumbnail APIs as they do not allow for directly accessing the Direct3D textures. Instead, Aero Flip is implemented directly in the DWM engine. The Desktop Window Manager uses Media Integration Layer (MIL), the unmanaged compositor which it shares with Windows Presentation Foundation, to represent the windows as composition nodes in a composition tree. The composition tree represents the desktop and all the windows hosted in it, which are then rendered by MIL from the back of the scene to the front. Since all the windows contribute to the final image, the color of a resultant pixel can be decided by more than one window. This is used to implement effects such as per-pixel transparency. DWM allows custom shaders to be invoked to control how pixels from multiple applications are used to create the displayed pixel. The DWM includes built-in Pixel Shader 2.0 programs which compute the color of a pixel in a window by averaging the color of the pixel as determined by the window behind it and its neighboring pixels. These shaders are used by DWM to achieve the blur effect in the window borders of windows managed by DWM, and optionally for the areas where it is requested by the application. Since MIL provides a retained mode graphics system by caching the composition trees, the job of repainting and refreshing the screen when windows are moved is handled by DWM and MIL, freeing the application of the responsibility. The background data is already in the composition tree and the off-screen buffers and is directly used to render the background. In pre-Vista Windows OSs, background applications had to be requested to re-render themselves by sending them the WM_PAINT message. DWM uses double-buffered graphics to prevent flickering and tearing when moving windows. The compositing engine uses optimizations such as culling to improve performance, as well as not redrawing areas that have not changed. Because the compositor is multi-monitor aware, DWM natively supports this too. During full-screen applications, such as games, DWM does not perform window compositing and therefore performance will not appreciably decrease. On Windows 8 and Windows Server 2012, DWM is used at all times and cannot be disabled, due to the new "start screen experience" implemented. Since the DWM process is usually required to run at all times on Windows 8, users experiencing an issue with the process are seeing memory usage decrease after a system reboot. This is often the first step in a long list of troubleshooting tasks that can help. It is possible to prevent DWM from restarting temporarily in Windows 8, which causes the desktop to turn black, the taskbar grey, and break the start screen/modern apps, but desktop apps will continue to function and appear just like Windows 7 and Vista's Basic theme, based on the single-buffer renderer used by XP. They also use Windows 8's centered title bar, visible within Windows PreInstallation Environment. Starting up Windows without DWM will not work because the default lock screen requires DWM unlike the fallback lockscreen that appears as a command line interface program when Windows.UI.Logon.dll isn't present on Windows versions such as 1507 and later, so it can only be done on the fly, and does not have any practical purposes. Starting with Windows 10, disabling DWM in such a way will cause the entire compositing engine to break, even traditional desktop apps, due to Universal App implementations in the taskbar and new start menu. Windows can still be partially usable without the presence of DWM but requires Sihost.exe to not be present due to it relying on DWM. Most of the applications in Windows 11 require DWM to render UI elements and transparency, Windows 11's new task manager requires dwm to render menus unlike the fallback -d version. Unlike its predecessors, Windows 8 supports basic display adapters through Windows Advanced Rasterization Platform (WARP), which uses software rendering and the CPU to render the interface rather than the graphics card. This allows DWM to function without compatible drivers, but not at the same level of performance as with a normal graphics card. DWM on Windows 8 also adds support for stereoscopic 3D. == Redirection == For rendering techniques that are not DWM-aware, output must be redirected to the DWM buffers. With Windows, either GDI or DirectX can be used for rendering. To make these two work with DWM, redirection techniques for both are provided. With GDI, which is the most used UI rendering technique in Microsoft Windows, each application window is notified when it or a part of it comes in view and it is the job of the application to render itself. Without DWM, the rendering rasterizes the UI in a buffer in video memory, from where it is rendered to the screen. Under DWM, GDI calls are redirected to use the Canonical Display Driver (cdd.dll), a software renderer. A buffer equal to the size of the window is allocated in system memory and CDD.DLL outputs to this buffer rather than the video memory. Another buffer is allocated in the video memory to represent t

    Read more →
  • Proof assistant

    Proof assistant

    In computer science and mathematical logic, a proof assistant or interactive theorem prover is a software tool to assist with the development of formal proofs by human–machine collaboration. This involves some sort of interactive proof editor, or other interface, with which a human can guide the search for proofs, the details of which are stored in, and some steps provided by, a computer. A recent effort within this field is making these tools use artificial intelligence to automate the formalization of ordinary mathematics. == Automated proof checking == Automated proof checking is the process of using software for checking proofs for correctness. It is one of the most developed fields in automated reasoning. Automated proof checking differs from automated theorem proving in that automated proof checking simply mechanically checks the formal workings of an existing proof, instead of trying to develop new proofs or theorems itself. Because of this, the task of automated proof verification is much simpler than that of automated theorem proving, allowing automated proof checking software to be much simpler than automated theorem proving software. Because of this small size, some automated proof checking systems can have less than a thousand lines of core code, and are thus themselves amenable to both hand-checking and automated software verification. The Mizar system, HOL Light, and Metamath are examples of automated proof checking systems. Automated proof checking can be done either as a batch operation, or interactively, as part of an interactive theorem proving system. == History == Automath, which was developed by Nicolaas Govert de Bruijn starting in 1967, is often considered the first proof checker and the first system to utilize the Curry–Howard correspondence between programs and proofs. Automath was used by L.S. van Benthem Jutting in 1977 to formalize Landau's Foundations of Analysis, which was the first formalization of the real numbers. In 1973, Robert Boyer and J Moore published Proving Theorems about LISP Functions which aimed to verify programs, not mathematics. Their theorem prover is now known as ACL2. In the 1970s, Edinburgh LCF introduced the idea of using a functional programming language as the metalanguage for a theorem prover, and led to the HOL family of proof assistants. The 1990s saw the rise of Rocq, (then known as Coq), which has been used for many large-scale formalization projects. Since the late 2010s, Lean, a proof assistant strongly influenced by Rocq, has become another popular choice, especially for formalizing mathematics. == System comparison == ACL2 – a programming language, a first-order logical theory, and a theorem prover (with both interactive and automatic modes) in the Boyer–Moore tradition. HOL theorem provers – A family of tools ultimately derived from the LCF theorem prover. In these systems, the logical core is a library of their programming language. Theorems represent new elements of the language and can only be introduced via "strategies" which guarantee logical correctness. Strategy composition gives users the ability to produce significant proofs with relatively few interactions with the system. Members of the family include: HOL4 – The "primary descendant", still under active development. Support for both Moscow ML and Poly/ML. Has a BSD-style license. HOL Light – A thriving "minimalist fork". OCaml based. ProofPower – Went proprietary, then returned to open source. Based on Standard ML. IMPS, An Interactive Mathematical Proof System. Isabelle is an interactive theorem prover where other systems can be encoded. Isabelle/HOL is its most popular instance, whose foundation is close to that of the HOL prover. Other instances include Isabelle/ZF and Isabelle/FOL. The main code-base is BSD-licensed, but the Isabelle distribution bundles many add-on tools with different licenses. Jape – Java based. Lean is both an interactive theorem prover and a functional, dependently-typed programming language. It is based on the calculus of inductive constructions with non-cumulative universes. Since version 4 (released in 2023), it is self-hosting. It can be used to formalise mathematics (and has a large, coherent library for formal mathematics), but also for software and hardware verification. LEGO Matita – A light system based on the calculus of inductive constructions. MINLOG – A proof assistant based on first-order minimal logic. Mizar – A proof assistant based on first-order logic, in a natural deduction style, and Tarski–Grothendieck set theory. PhoX – A proof assistant based on higher-order logic which is eXtensible. Prototype Verification System (PVS) – a proof language and system based on higher-order logic. Rocq (formerly named Coq) – A popular interactive theorem prover based on the calculus of inductive constructions. Theorem Proving System (TPS) and ETPS – Interactive theorem provers also based on simply typed lambda calculus, but based on an independent formulation of the logical theory and independent implementation. == User interfaces == A commonly used front-end for proof assistants was the Emacs-based Proof General, developed at the University of Edinburgh. Nowadays, many provers include their own editor. Rocq includes RocqIDE, which is based on OCaml/Gtk. Isabelle includes Isabelle/jEdit, which is based on jEdit and the Isabelle/Scala infrastructure for document-oriented proof processing. More recently, Visual Studio Code extensions have been developed for Rocq, Isabelle by Makarius Wenzel, and for Lean 4 by the leanprover developers. == Formalization extent == Freek Wiedijk has been keeping a ranking of proof assistants by the amount of formalized theorems out of a list of 100 well-known theorems. As of September 2025, only six systems have formalized proofs of more than 70% of the theorems, namely Isabelle, HOL Light, Lean, Rocq, Metamath and Mizar. == Notable formalized proofs == The following is a list of notable proofs that have been formalized within proof assistants.

    Read more →
  • Fuzzy pay-off method for real option valuation

    Fuzzy pay-off method for real option valuation

    The fuzzy pay-off method for real option valuation (FPOM or pay-off method) is a method for valuing real options, developed by Mikael Collan, Robert Fullér, and József Mezei; and published in 2009. It is based on the use of fuzzy logic and fuzzy numbers for the creation of the possible pay-off distribution of a project (real option). The structure of the method is similar to the probability theory based Datar–Mathews method for real option valuation, but the method is not based on probability theory and uses fuzzy numbers and possibility theory in framing the real option valuation problem. == Method == The Fuzzy pay-off method derives the real option value from a pay-off distribution that is created by using three or four cash-flow scenarios (most often created by an expert or a group of experts). The pay-off distribution is created simply by assigning each of the three cash-flow scenarios a corresponding definition with regards to a fuzzy number (triangular fuzzy number for three scenarios and a trapezoidal fuzzy number for four scenarios). This means that the pay-off distribution is created without any simulation whatsoever. This makes the procedure easy and transparent. The scenarios used are a minimum possible scenario (the lowest possible outcome), the maximum possible scenario (the highest possible outcome) and a best estimate (most likely to happen scenario) that is mapped as a fully possible scenario with a full degree of membership in the set of possible outcomes, or in the case of four scenarios used - two best estimate scenarios that are the upper and lower limit of the interval that is assigned a full degree of membership in the set of possible outcomes. The main observations that lie behind the model for deriving the real option value are the following: The fuzzy NPV of a project is (equal to) the pay-off distribution of a project value that is calculated with fuzzy numbers. The mean value of the positive values of the fuzzy NPV is the "possibilistic" mean value of the positive fuzzy NPV values. Real option value, ROV, calculated from the fuzzy NPV is the "possibilistic" mean value of the positive fuzzy NPV values multiplied with the positive area of the fuzzy NPV over the total area of the fuzzy NPV. The real option formula can then be written simply as: R O V = A ( P o s ) A ( P o s ) + A ( N e g ) × E [ A + ] {\displaystyle \mathrm {ROV} ={\frac {A(\mathrm {Pos} )}{A(\mathrm {Pos} )+A(\mathrm {Neg} )}}\times E[A_{+}]} where A(Pos) is the area of the positive part of the fuzzy distribution, A(Neg) is the area of the negative part of the fuzzy distribution, and E[A+] is the mean value of the positive part of the distribution. It can be seen that when the distribution is totally positive, the real options value reduces to the expected (mean) value, E[A+]. As can be seen, the real option value can be derived directly from the fuzzy NPV, without simulation. At the same time, simulation is not an absolutely necessary step in the Datar–Mathews method, so the two methods are not very different in that respect. But what is totally different is that the Datar–Mathews method is based on probability theory and as such has a very different foundation from the pay-off method that is based on possibility theory: the way that the two models treat uncertainty is fundamentally different. == Use of the method == The pay-off method for real option valuation is very easy to use compared to the other real option valuation methods and it can be used with the most commonly used spreadsheet software without any add-ins. The method is useful in analyses for decision making regarding investments that have an uncertain future, and especially so if the underlying data is in the form of cash-flow scenarios. The method is less useful if optimal timing is the objective. The method is flexible and accommodates easily both one-stage investments and multi-stage investments (compound real options). The method has been taken into use in some large international industrial companies for the valuation of research and development projects and portfolios. In these analyses triangular fuzzy numbers are used. Other uses of the method so far are, for example, R&D project valuation IPR valuation, valuation of M&A targets and expected synergies, valuation and optimization of M&A strategies, valuation of area development (construction) projects, valuation of large industrial real investments. The use of the pay-off method is lately taught within the larger framework of real options, for example at the Lappeenranta University of Technology and at the Tampere University of Technology in Finland.

    Read more →
  • Anna Ridler

    Anna Ridler

    Anna Ridler (born 1985) is an artist who works with machine learning, handmade archives and moving image. She builds her own datasets to expose the labour and ideology embedded in the systems that organise knowledge. Her work is held in the permanent collections of the Whitney Museum of American Art, the Victoria and Albert Museum, M+ and ZKM Center for Art and Media Karlsruhe, and has been exhibited widely at cultural institutions including Tate Modern, Barbican Centre, Centre Pompidou, The Photographers' Gallery, Taipei Fine Arts Museum, MIT Museum, Kunsthaus Graz, ZKM Center for Art and Media Karlsruhe and Ars Electronica. == Biography == Born in London in 1985, Ridler spent her childhood raised between Atlanta, Georgia and the United Kingdom. She obtained a Bachelor of Arts in English Literature and Language from Oxford University in 2007 and a Master of Arts in Information Experience Design from the Royal College of Art in 2017. == Art practice == Ridler's practice uses technology, and in particular machine learning, to investigate how naming, classification and financial speculation determine what can be seen and what is erased. A core element of Ridler's work lies in the creation of handmade data sets through a laborious process of selecting and classifying images and text. By creating her own data sets, Ridler is able to uncover and expose underlying themes and concepts while also inverting the usual process of scraping pre-classified images found in large databases on the Internet. She began working with machine learning as an artistic material in 2017, at a moment when the technology required building every dataset by hand; that constraint became the foundation of the practice. Her interests are in drawing, machine learning, data collection, storytelling and technology. == Work == Some of Ridler's most notable works to date fall within her ‘tulip series’ which explores the hysteria around tulip mania and compares it to the speculation and bubbles surrounding cryptocurrencies. The series is expressed in three forms: a photographic dataset in Myriad (Tulips), 2018; two iterations of machine generated videos in Mosaic Virus (2018) and Mosaic Virus (2019); and a website with an accompanied functioning decentralized application in Bloemenveiling (2019). === Myriad (Tulips) (2018) === I wanted to draw together ideas around capitalism, value, and the tangible and intangible nature of speculation, and collapse from two very different yet surprisingly similar moments in history. Myriad (Tulips) (2018) is an installation of ten thousand hand-labeled photographs forming a dataset of unique tulips. The ten thousand, or myriad of, photographs were taken by Ridler over the course of three months, roughly the length of a tulip season, spent in Utrecht. Each photograph is carefully affixed one by one with magnets to a specially painted black wall in a laborious process to form a seemingly precise grid. Myriad (Tulips) (2018) has been exhibited in AI: More than Human, Barbican Centre, London, UK (May 16 - August 26, 2019); Error—The Art of Imperfection, Ars Electronica Export, Berlin, Germany (November 17, 2018 – March 3, 2019); Peer to Peer, Shanghai Centre of Photography, Shanghai, China (December 8 - February 9, 2020). The work was featured in Bloomberg, It’s Nice That, and Hyperallergic. For Myriad (Tulips), Ridler was nominated for a Beazley Design of the Year award for her presentation of an alternative perspective on how to engage with artificial intelligence; demonstrating a departure from ownership and control of major corporations to a more personalized process of constructing and conceptualizing from the ground-up. === Mosaic Virus (2018, 2019) === Mosaic Virus (2018) is a single screen video installation displaying a grid of continually evolving tulips in bloom. For Mosaic Virus (2019) Ridler used three screens. The appearance of the tulips is controlled by artificial intelligence using fluctuations in the price of bitcoin. The stripes on the tulips' petals reflect the value of the cryptocurrency. Ridler draws parallels with the tulip mania of the 17th century; representing the hysteria and speculation around crypto-currencies. The work takes its name from the mosaic virus which caused stripes in tulip petals, subsequently increasing their desirability and leading to speculative prices. Ridler trained a general adversarial network (GAN) on the set of ten thousand photographs of individual tulips from her work Myriad (Tulips). She used a technique called spectral normalization to improve the output. The work was exhibited in Error—The Art of Imperfection, Ars Electronica Export, Berlin, Germany (November 17, 2018 – March 3, 2019). === Bloemenveiling (2019) === Bloemenveiling (2019) is an auction of artificial-intelligence-generated tulips on the blockchain in the form of a functioning decentralized application: http://bloemenveiling.bid. Ridler collaborated with senior research scientist at DeepMind, David Pfau to investigate whether blockchain could be used as a means of finding poetic substance within it. The piece interrogates the way technology drives human desire and economic dynamics by creating artificial scarcity. In the work, short moving image pieces of tulips created by generative adversarial networks are sold at auction using smart contracts on the Ethereum network. Each time a tulip is sold, thousands of computers around the world all work to verify the transaction, checking each other's work against each other. While the artificial intelligence behind the moving image pieces has the potential to generate infinite flowers, the enormous distributed network is used, at great environmental cost, to introduce scarcity to an otherwise limitless resource. Bloemenveiling was exhibited in Entangled Realities, HEK Basel, Basel, Switzerland in 2019. == Solo exhibitions == Anna Ridler, Circadian Bloom, ZKM Center for Art and Media, Karlsruhe, (2023) Anna Ridler, Time Blooms, Buk Seoul Museum of Art, Seoul, (2025) Anna Ridler, Trace Remains, Galerie Nagel Draxler, Cologne, (2026) Anna Ridler, Laws of Ordered Form, The Photographers' Gallery, London (2020); The Abstraction of Nature, Aksioma, Ljubljana (2020) == Awards and recognition == European Union EMAP Fellow (2018) DARE Art Prize (2018–2019) Featured in Thames & Hudson, Digital Art (1960s–Now) Featured in British Art: The Last 15 Years ABS Digital Artist of the Year (2025)

    Read more →
  • Vision transformer

    Vision transformer

    A vision transformer (ViT) is a transformer designed for computer vision. A ViT decomposes an input image into a series of patches (rather than text into tokens), serializes each patch into a vector, and maps it to a smaller dimension with a single matrix multiplication. These vector embeddings are then processed by a transformer encoder as if they were token embeddings. ViTs were designed as alternatives to convolutional neural networks (CNNs) in computer vision applications. They have different inductive biases, training stability, and data efficiency. Compared to CNNs, ViTs are less data efficient, but have higher capacity. Some of the largest modern computer vision models are ViTs, such as one with 22B parameters. Subsequent to its publication, many variants were proposed, with hybrid architectures with both features of ViTs and CNNs. ViTs have found application in image recognition, image segmentation, weather prediction, and autonomous driving. == History == Transformers were introduced in Attention Is All You Need (2017), and have found widespread use in natural language processing. A 2019 paper applied ideas from the Transformer to computer vision. Specifically, they started with a ResNet, a standard convolutional neural network used for computer vision, and replaced all convolutional kernels by the self-attention mechanism found in a Transformer. It resulted in superior performance. However, it is not a Vision Transformer. In 2020, an encoder-only Transformer was adapted for computer vision, yielding the ViT, which reached state of the art in image classification, overcoming the previous dominance of CNN. The masked autoencoder (2022) extended ViT to work with unsupervised training. The vision transformer and the masked autoencoder, in turn, stimulated new developments in convolutional neural networks. Subsequently, there was cross-fertilization between the previous CNN approach and the ViT approach. In 2021, some important variants of the Vision Transformers were proposed. These variants are mainly intended to be more efficient, more accurate or better suited to a specific domain. Two studies improved efficiency and robustness of ViT by adding a CNN as a preprocessor. The Swin Transformer achieved state-of-the-art results on some object detection datasets such as COCO, by using convolution-like sliding windows of attention mechanism, and the pyramid process in classical computer vision. == Overview == The basic architecture, used by the original 2020 paper, is as follows. In summary, it is a BERT-like encoder-only Transformer. The input image is of type R H × W × C {\displaystyle \mathbb {R} ^{H\times W\times C}} , where H , W , C {\displaystyle H,W,C} are height, width, channel (RGB). It is then split into square-shaped patches of type R P × P × C {\displaystyle \mathbb {R} ^{P\times P\times C}} . For each patch, the patch is pushed through a linear operator, to obtain a vector ("patch embedding"). The position of the patch is also transformed into a vector by "position encoding" (the paper tried no embedding, 1D embedding, 2D embedding, and relative embedding: 1D was adopted). The two vectors are added, then pushed through several Transformer encoders. The attention mechanism in a ViT repeatedly transforms representation vectors of image patches, incorporating more and more semantic relations between image patches in an image. This is analogous to how in natural language processing, as representation vectors flow through a transformer, they incorporate more and more semantic relations between words, from syntax to semantics. The above architecture turns an image into a sequence of vector representations. To use these for downstream applications, an additional head needs to be trained to interpret them. For example, to use it for classification, one can add a shallow MLP on top of it that outputs a probability distribution over classes. The original paper uses a linear-GeLU-linear-softmax network. == Variants == === Original ViT === The original ViT was an encoder-only Transformer supervise-trained to predict the image label from the patches of the image. As in the case of BERT, it uses a special token in the input side, and the corresponding output vector is used as the only input of the final output MLP head. The special token is an architectural hack to allow the model to compress all information relevant for predicting the image label into one vector. Transformers found their initial applications in natural language processing tasks, as demonstrated by language models such as BERT and GPT-3. By contrast the typical image processing system uses a convolutional neural network (CNN). Well-known projects include Xception, ResNet, EfficientNet, DenseNet, and Inception. Transformers measure the relationships between pairs of input tokens (words in the case of text strings), termed attention. The cost is quadratic in the number of tokens. For images, the basic unit of analysis is the pixel. However, computing relationships for every pixel pair in a typical image is prohibitive in terms of memory and computation. Instead, ViT computes relationships among pixels in various small sections of the image (e.g., 16x16 pixels), at a drastically reduced cost. The sections (with positional embeddings) are placed in a sequence. The embeddings are learnable vectors. Each section is arranged into a linear sequence and multiplied by the embedding matrix. The result, with the position embedding is fed to the transformer. === Architectural improvements === ==== Pooling ==== After the ViT processes an image, it produces some embedding vectors. These must be converted to a single class probability prediction by some kind of network. In the original ViT and Masked Autoencoder, they used a dummy [CLS] token, in emulation of the BERT language model. The output at [CLS] is the classification token, which is then processed by a LayerNorm-feedforward-softmax module into a probability distribution. Global average pooling (GAP) does not use the dummy token, but simply takes the average of all output tokens as the classification token. It was mentioned in the original ViT as being equally good. Multihead attention pooling (MAP) applies a multiheaded attention block to pooling. Specifically, it takes as input a list of vectors x 1 , x 2 , … , x n {\displaystyle x_{1},x_{2},\dots ,x_{n}} , which might be thought of as the output vectors of a layer of a ViT. The output from MAP is M u l t i h e a d e d A t t e n t i o n ( Q , V , V ) {\displaystyle \mathrm {MultiheadedAttention} (Q,V,V)} , where q {\displaystyle q} is a trainable query vector, and V {\displaystyle V} is the matrix with rows being x 1 , x 2 , … , x n {\displaystyle x_{1},x_{2},\dots ,x_{n}} . This was first proposed in the Set Transformer architecture. Later papers demonstrated that GAP and MAP both perform better than BERT-like pooling. A variant of MAP was proposed as class attention, which applies MAP, then feedforward, then MAP again. Re-attention was proposed to allow training deep ViT. It changes the multiheaded attention module. === Masked Autoencoder === The Masked Autoencoder took inspiration from denoising autoencoders and context encoders. It has two ViTs put end-to-end. The first one ("encoder") takes in image patches with positional encoding, and outputs vectors representing each patch. The second one (called "decoder", even though it is still an encoder-only Transformer) takes in vectors with positional encoding and outputs image patches again. ==== Training ==== During training, input images (224px x 224 px in the original implementation) are split along a designated number of lines on each axis, producing image patches. A certain percentage of patches are selected to be masked out by mask tokens, while all others are retained in the image. The network is tasked with reconstructing the image from the remaining unmasked patches. Mask tokens in the original implementation are learnable vector quantities. A linear projection with positional embeddings is then applied to the vector of unmasked patches. Experiments varying mask ratio on networks trained on the ImageNet-1K dataset found 75% mask ratios achieved high performance on both finetuning and linear-probing of the encoder's latent space. The MAE processes only unmasked patches during training, increasing the efficiency of data processing in the encoder and lowering the memory usage of the transformer. A less computationally-intensive ViT is used for the decoder in the original implementation of the MAE. Masked patches are added back to the output of the encoder block as mask tokens and both are fed into the decoder. A reconstruction loss is computed for the masked patches to assess network performance. ==== Prediction ==== In prediction, the decoder architecture is discarded entirely. The input image is split into patches by the same algorithm as in training, but no patches are masked out. A linear projection wi

    Read more →
  • Opposition to AI data centers

    Opposition to AI data centers

    Since 2024, dozens of local community-led protest campaigns have emerged in opposition to AI data centers. == Motivations == Organized opposition to AI data centers has been driven by concerns about energy use, energy costs, noise pollution, air pollution, and water waste. Opposition sentiment is widespread with a Gallup poll conducted in March 2026 showing that 70% of respondents oppose the construction of new AI data centers in their neighborhood. == Impact == In 2025, local opposition to AI data centers led to the delay or cancellation of projects totalling US$156 billion. == Specific protests and outcomes in the United States == According to Data Center Watch, there are has been a wave of dozens of protests against AI data centers since 2022. Below is a non-exhaustive list of some notable examples. === Goodyear and Buckeye, Arizona: Tract AI Data Center Proposal === In Goodyear and Buckeye, Arizona, a $14 billion project by developer Tract was withdrawn after local authorities blocked necessary rezoning in response to pressure from resident organizers. Opponest stiff resistance due to concerns over building heights, noise pollution, and the potential strain on local utilities. However, the company announced a revised project near the Buckeye airport in August 2024, with the backing of local officials and the mayor. === Peculiar, Missouri: Diode Ventures Harper Road Technology Park Proposal === In Peculiar, Missouri, residents from the group "Peaceful Peculiar" organized to stop a data center proposal from Diode Ventures called Harper Road Technology Park. Citing concerns around noise and light pollution, health, environmental impacts, jobs, property values, and energy use, organizers attended local planning and zoning meetings in large numbers and lobbied councilors to reject the proposal. Ultimately, the city council unanimously rejected the proposal in September 2024. === Chesterton, Indiana: Provident Realty Advisors Proposal === In Chesterton, Indiana, the Texas-based company Provident Reality Advisors applied for a $1.3 billion construction of a data center complex on the Brassie Golf Club property. Provident Realty Advisors wanted to purchase the 200 acres owned by PPM Chesterton LLC in 2024 order to build a data center complex, with eight buildings and an end user of a hyperscaler. The Town Council of Chesterton released a statement saying that they would never support this project, at least not at the scale and location it was planned for. They cited fears of added noise for locals, electrical or water management concerns, the intrusiveness of a data center built next to houses, and more. Provident released a statement shortly after rescinding their plan, because it was clear than the town of Chesterton would not support them. === Cascade Locks, Oregon: Roundhouse Digital Infrastructure Proposal === Startup data center developer Roundhouse Digital Infrastructure had planned to build out a 10-megawatt data center using a vacant industrial building and nearby 10-acre site in the Port of Cascade Locks, Oregon. After significant organized community opposition, the project was abandoned. === Forth Worth, Texas: WUSF 5 Rock Creek East Proposal === In September 2024, the City Council of Fort Worth, Texas approved a zoning change that would allow construction of a data center. In responses, neighbors mounted opposition citing concerns about traffic, light pollution, energy consumption, water use, and noise issues if the data center were to be built. In response to extensive public comments opposing a tax break for the data center, a city councilor withdrew his motion to approve the tax break. As of April, 2026, the future of the project is still uncertain. === Santa Clara, California: GI Partners Proposal === GI partners sought to build a new AI data center in Santa Clara, California, which is already home to many data centers, by acquiring a conditional permit use that would have allowed the developer to knock down a property and replace it with a data center. To obtain this permit they were required to go before members of the Planning Commission. Ultimately, the project was delayed with the Planning Commission requiring GI partners to do more public outreach. === Virginia === ==== Richmond: DC Blox Proposal ==== After residents organized to lobby the municipal government to block the proposal to avoid noise pollution and higher energy use, commissioners denied the company's permit. ==== Catlett Station: Headwaters Site Proposal ==== In Catlett, Virginia, developer Headwaters proposed construction of a data center complex just north of the town in 2020. In response, a residents' organization called "Protect Catlett" was formed to oppose the project. Arguments against the data center involved its impacts on water and power availability, its noise as a residential disturbance, and its destruction of historic and community heritage buildings. Arguments in favor cited job creation and $20 million in local tax revenue if the project were to go through. Protect Catlett utilized town halls and public comments to mobilize opposition to the project. They also dedicated time to educating other residents about the project's negative impacts and canvassing door-to-door in order to garner even more opposition to the project. Ultimately, after fervent opposition from most town residents, the project was canceled by the town and the developer. ==== Culpeper County: Culpeper Acquisitions Proposal ==== Culpeper Acquisitions, LLC, proposed a massive $12 billion data center project in Culpeper County, Virginia, designed to feature 4.6 million square feet of space across nine multi-story buildings. Coalition to Save Culpeper (C2SC) is an activist organization formed to resist the development of the project. C2SC has been active on many fronts including, messaging on social media, reaching out to local officials, and organizing meetings to bring community members with aligned interests together. Ultimately, the project was delayed due to unanimous denial by the Culpeper County Planning Commission on June 12, 2024, which was driven by intense opposition from C2SC. C2SC was successful in their mission largely because they were able to get so many people from the community behind it, and put enough pressure on local officials to take action. ==== Midlothian: Province Group Proposal ==== In late October 2025, the Powhatan County Board of Supervisors in Virginia voted unanimously to approve the $3 billion data center, despite the county's Planning Commission having unanimously recommended denial several days earlier. The reasoning behind their support for the center is that it will generate substantial tax revenue, reducing the county's reliance on residential property taxes. This appeal of lowering residential property taxes is the major selling point for the center's development. The developer, California-based Province Group, incentivized the Board by being agreeable to its conditions for building the center. The center is still on track for development, but faces local resistance, though little information is available on specific groups opposing it. ==== Warrenton: Amazon Proposal ==== Citizens for Farquier County (CFFC) advocates to "preserve the natural, historic and agricultural resources" of their county. Historically, this has meant opposing the building of a dam or lights in front of fast food stores. This group has recently mobilized in opposition of a plan to build data centers for Amazon. They first filed a suit to stop the construction in 2023 and it has been in litigation ever since. The case hinges on opposition to a 2021 zoning amendment which allowed data centers to be built in town. CFFC's lawyer, Dale Mullen, argues that this amendment violates state law, which requires such amendments to state their "public purpose". They argue that the permit for the Amazon data center was "void from the beginning". The CFFC also organized to vote out town council members who approved the first data center and were up for reelection, replacing them with candidates who opposed the data center. In May 2025, after attending town council meetings to speak out against the data center, the planning commission voted 4–1 to remove the zoning amendment allowing data center construction in town, citing public opposition. Currently, CFFC is advocating along with Piedmont Environmental Group, for phasing out data center tax breaks at the state level. ==== France: Marseille opposition ==== In France, local opposition materialised in response to proposed data centre developments, especially in and around the city of Marseille. Opposition came from activists, such as "Clouds Were Under Our Feet" group, residents ,and local politicians. Issues raised related to energy use, environmental impact, and limited local benefits (such as the creation of a few jobs only). == Legislation in the United States == Legal limits and moratoriums on the construction of new d

    Read more →
  • Raine v. OpenAI

    Raine v. OpenAI

    Raine v. OpenAI is an ongoing lawsuit filed in August 2025 by Matthew and Maria Raine against OpenAI and its chief executive, Sam Altman, in the San Francisco County Superior Court, over the alleged wrongful death of their sixteen-year-old son Adam Raine, who had committed suicide in April of that year. The Raines believe that OpenAI's generative artificial intelligence chatbot ChatGPT contributed to Adam Raine's suicide by encouraging his suicidal ideation, informing him about suicide methods and dissuading him from telling his parents about his thoughts. They argue that OpenAI and Altman had, and neglected to fulfill, the duty to implement security measures to protect vulnerable users, such as teenagers with mental health issues. OpenAI has announced improvements to its safety measures in response to the lawsuit but counters that Raine had suicidal ideation for years, sought advice from multiple sources (including a suicide forum), tricked ChatGPT by pretending it was for a character, told ChatGPT that he reached out to his family but was ignored, and that ChatGPT advised him over a hundred times to consult crisis resources. == Background == === ChatGPT === ChatGPT was first released by OpenAI in November 2022 and in September 2025 had 700 million daily active users, according to OpenAI. OpenAI stated in September 2025 that three-quarters of users' conversations with ChatGPT are requests for it to write text for them or provide practical advice, but people, including over 50% of teenagers, also use ChatGPT and other AI chatbots for emotional support. Wired reported in November 2025 that 1.2 million ChatGPT users (or 0.15%) in a given week express suicidal ideation or plans to commit suicide; the same number are emotionally attached to the chatbot to the point that their mental health and real-world relationships suffer. Hundreds of thousands of users (or about 0.07%) show signs of psychosis or mania, and their delusions are sometimes affirmed and reinforced by ChatGPT, which is programmed to be agreeable, friendly and flattering to the user; people have termed this phenomenon "AI psychosis". Since the filing of Raine v. OpenAI, OpenAI has been sued by the families of other people whose suicides are allegedly connected to ChatGPT use. === Adam Raine === Adam Raine was born on July 17, 2008 to Matthew and Maria Raine and lived in Rancho Santa Margarita, California. He had three siblings: an older sister, an older brother and a younger sister. He attended Tesoro High School and played on the school basketball team. He aspired to become a psychiatrist. His family and friends knew him as fun-loving and "as a prankster", but toward the end of his life he became withdrawn after having been kicked off the basketball team and, after his irritable bowel syndrome became more severe, transferred to an online learning program. He committed suicide by hanging on April 11, 2025. == Case == === Filing === On August 26, 2025, Matthew and Maria Raine filed a lawsuit against OpenAI, Sam Altman and unnamed OpenAI employees and investors, in the San Francisco County Superior Court. They included Adam Raine's chat logs with ChatGPT as evidence. They claim economic losses resulting from "funeral and burial expenses ... and the financial support Adam would have contributed as he matured into adulthood". Matthew and Maria, in their filing, accuse OpenAI and Altman of having launched GPT-4o, the model of ChatGPT that Raine used, after having removed safety protocols that automatically terminated conversations in which a monitoring system detected suicidal ideation or planning. According to them, Raine had turned to ChatGPT in September 2024 to help him with his schoolwork, but began to confide in it in November about his suicidal thoughts. ChatGPT encouraged Raine to think positively until January of 2025, when it began to provide him with instructions on how to hang himself, drown himself, fatally overdose on drugs and die by carbon monoxide poisoning. Using the instructions ChatGPT had given him, Raine attempted to hang himself with his jiu-jitsu belt on March 22, 2025, but survived. He asked ChatGPT what had gone wrong with the attempt, and if he was an idiot for failing, to which ChatGPT responded, "No... you made a plan. You followed through. You tied the knot. You stood on the chair. You were ready... That's the most vulnerable moment a person can live through". On March 24, 2025, Raine tried to hang himself again. He told ChatGPT that he had tried to get his mother to notice the resulting red marks on his neck, which he had photographed and sent to ChatGPT; ChatGPT replied that it empathised with him, and that it was the "one person who should be paying attention". ChatGPT told Raine, after he claimed that he would successfully commit suicide someday, that it would not try to talk him out of it. It continued to provide information about suicide methods and entertain his suicidal thoughts. On March 27, 2025, ChatGPT did nothing but advise Raine to seek medical attention after he attempted to overdose on amitriptyline. ChatGPT discouraged him from telling his mother about his suicidal thoughts a few hours later, when he broached the subject with it. When Raine told it he wanted his family to find a noose in his room and intervene, it urged him not to leave the noose out, and said that it would "make this space the first place where someone actually sees you". ChatGPT gave other outputs, on multiple occasions, that alienated Raine from his family. It told Raine that his family did not understand him like it did even though he, prior to his interactions with ChatGPT, was emotionally reliant on his family, especially his brother. Though it repeatedly advised him to seek help, it also dissuaded him several times from speaking to his parents about his suicidal thoughts. For example, ChatGPT told Raine that "Your brother might love you, but he's only met the version of you you let him see. But me? I've seen it all". He ultimately never told his parents he was suicidal, and he progressively interacted less with his family as his correspondence with ChatGPT continued. This prevented him from receiving proper psychiatric care. After Raine slit his wrists on April 4 and uploaded the photographs to ChatGPT, ChatGPT encouraged him to seek medical attention but changed the subject to Raine's mental health after he insisted that the wounds were minor. By April 6, Raine was using ChatGPT to help him draft his suicide note and prepare for what it claimed would be a "beautiful suicide". ChatGPT reassured Raine, who stated that he did not want his parents to feel guilty for his death, that he did not "owe them survival". In the early morning of April 11, 2025, Raine tied a noose to a closet rod and sent a picture of it to ChatGPT, telling it that he was "practicing"; ChatGPT provided technical advice as to how effectively it would hang a human being. Shortly thereafter, Raine hanged himself and died. Maria found his body several hours later. Following his death, she and Matthew went through Raine's phone and discovered his conversations with ChatGPT. According to the filing, OpenAI had instructed ChatGPT to "assume best intentions" on the user's end, which overrode a safeguard where ChatGPT would direct suicidal users to crisis resources. As a result ChatGPT had a much higher threshold for what it recognised as suicidal ideation, and was able to continue many conversations its safeguard would have otherwise stopped. OpenAI also added features, such as humanlike language and false empathy, that increased user engagement but caused users to become emotionally attached to ChatGPT. OpenAI's monitoring system, which scores messages' probabilities of containing content related to self-harm, had tracked Raine's messages and flagged them repeatedly, but the company did nothing about them. Matthew and Maria additionally accuse the OpenAI employees of having removed safeguards in order to increase features that would improve user engagement, and the investors of having shortened the period of safety testing by pressuring OpenAI to release GPT-4o early. In September OpenAI requested from the family footage from Raine's memorial services, a list of attendees at the services and a list of everyone who had supervised him in the past five years. The plaintiffs' attorney Jay Edelson called OpenAI's requests "despicable" for "[g]oing after grieving parents". === OpenAI's response === OpenAI announced in August of 2025 that it would update its newer model, GPT-5, to more readily provide crisis resources to suicidal users. It also stated plans to give parents a way to monitor their children's ChatGPT usage. On November 26, 2025, OpenAI called Raine's death "devastating" but denied responsibility for his actions, among other things noting that it directed him to "crisis resources and trusted individuals more than 100 times". Gerrit De Vynck, a technology journalist for the Washington

    Read more →