AI Chatbot Meme

AI Chatbot Meme — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

Sensory, Inc.

Sensory, Inc. is an American company which develops software AI technologies for speech, sound and vision. It is based in Santa Clara, California. Sensory’s technologies have shipped in over three billion products from hundreds of leading consumer electronics manufacturers including AT&T, Hasbro, Huawei, Google, Amazon, Samsung, LG, Mattel, Motorola, Plantronics, GoPro, Sony, Tencent, Garmin, LG, Microsoft, Lenovo, and more. Sensory has over 60 issued patents covering speech recognition in consumer electronics, biometric authentication, sensor/speech combinations, wake word technology, and more. == History == Sensory, Inc. was founded in 1994, originally as Sensory Circuits, by Forrest Mozer, Mike Mozer and Todd Mozer. The three had also co-founded ESS Technology years earlier. In 1999 Sensory acquired Fluent Speech Technologies, which was formed and started by a group of professors out of the Oregon Graduate Institute (formerly OGI, now OHSU). Fluent Speech Technologies developed high performance embedded speech engines, the technology from this acquisition is now the core technology used throughout Sensory's chip and software line. === Company timeline === 1994 – Founded 1995 – Introduces the RSC 164 - first commercially successful speech recognition IC 1998 – Introduces first speaker verification IC 2000 – Acquires Oregon based Fluent-Speech Technologies 2002 – Acquires Texas Instruments line of speech output ICs (the SC series) 2007 – Introduces first Voice User Interface for Bluetooth silicon (CSR BC-5) - BlueGenie 2008 - Sensory and BlueAnt partner on the V1 - Revolutionary new Bluetooth headset with a voice user interface. First wearable to use a voice user interface for control and best-reviewed speech recognition product in history 2009 – Introduced world's smallest text to speech system (TTS) and Truly HandsfreeTM Triggers/ wake words. 2010 – Introduced the NLP-5x – First Natural Language Voice Processor and TrulyHandsfree wake words in SDKs for Android, iOS, Linux, and Windows. NLP5x used the first generation of TrulyHandsfree wake words with low power and enhanced accuracy. 2011 – Sensory partners with Google and Microsoft to enable TrulyHandsfree as a front end to Goog411 and Bing411 2012 – Partnered with Tensilica to offer ultra-low power TrulyHandsfree wake words; introduced Speaker Verification and Speaker Identification for mobile phones and other consumer electronics. 2012 - TrulyHandsfree released into Samsung's Galaxy S2 for "Hey Galaxy" wake word 2013 – TrulyHandsfree wake words migrated to many new platforms and began shipping as MotoVoice in the Google-owned MotoX. Sensory's TrulyHandsfree in mobile takes off with the Galaxy S3 and S4 and Galaxy Note and is licensed into wearables like Google Glass. 2014 – Announced new initiative in Vision; added LG and Motorola as customers; received the 2014 Global Mobile Award for Best Mobile Technology Breakthrough at the GSMA Mobile World Congress in Barcelona, Spain (judges commented, "A big advance for the wearables market, this offers many benefits for consumers, increasing uptake and usage of many mobile apps, driving revenue for operators and content providers.") 2015-2018 - Licensed Google, Amazon, MSFT, Baidu, Huawei, ZTE, and many others with TrulyHandsfree wake words. Sensory develops first wake words for OK Google, Hey Siri, and Hey Cortana. 2019 - Sensory launched two new solutions: SoundID, sound identification, and TrulyNatural, embedded large vocabulary speech recognition. Sensory also acquired Vocalize.ai, an independent testing lab. 2020 - Sensory introduced VoiceHub, which allows the automated generation of wake words. 2021 - Sensory expands VoiceHub with speech recognition and NLU capabilities. The company initiated a new cloud platform, SensoryCloud.ai. 2022-Sensory rolls out SensoryCloud.ai with speech to text, text to speech, face & voice biometrics 2024- Sensory Automotive & TrulyNatural Speech-to-text On-Device launched == Technology and products == Sensory originally developed both hardware (Integrated Circuit - IC or "chip") and software platforms but migrated to software only around 2005 and added cloud and hybrid computing capabilities in 2021. Sensory's RSC-164 IC (Integrated Circuit or "chip") was used on NASA's Mars Polar Lander in the Mars Microphone on the Lander. Speech Synthesis SC-6x chips – acquired some speech synthesis technology from Texas Instruments. Sensory’s embedded AI solutions include the following: TrulyHandsfree (THF) - wake word detection and phrase spotting. TrulyNatural (TNL) - large vocabulary continuous speech recognition with NLU. TrulySecure (TS) - face and voice biometrics. TrulySecureSpeakerVerification (TSSV) - speaker and sound identification. VoiceHub - Online portal for creating custom wake words and speech recognition models with NLU. Sensory Automotive- Sensory Automotive is a full voice and vision suite of AI technologies that operate efficiently in the car without connecting to a network. The cloud initiative, SensoryCloud.ai, is targeting Speech To Text (STT), Text To Speech (TTS), Wake Word verification, face and voice recognition, and sound identification.
Read more →
Netsukuku

Netsukuku is an experimental peer-to-peer routing system, developed by the FreakNet MediaLab in 2005, created to build up a distributed network, anonymous and censorship-free, fully independent but not necessarily separated from the Internet, without the support of any server, Internet service provider and no central authority. Netsukuku is designed to handle up to 2128 nodes without any servers or central systems, with minimal CPU and memory resources. This mesh network can be built using existing network infrastructure components such as Wi-Fi. The project has been in slow development since 2005, never abandoning a beta state. It has also never been tested on large scale. == Operation == As of December 2011, the latest theoretical work on Netsukuku could be found in the author's master thesis Scalable Mesh Networks and the Address Space Balancing problem. The following description takes into account only the basic concepts of the theory. Netsukuku uses a custom routing protocol called QSPN (Quantum Shortest Path Netsukuku) that strives to be efficient and not taxing on the computational capabilities of each node. The current version of the protocol is QSPNv2. It adopts a hierarchical structure. 256 nodes are grouped inside a gnode (group node), 256 gnodes are grouped in a single ggnode (group of group nodes), 256 ggnodes are grouped in a single gggnode, and so on. This offers a set of advantages main documentation. The protocol relies on the fact that the nodes are not mobile and that the network structure does not change quickly, as several minutes may be required before a change in the network is propagated. However, a node that joins the network is immediately able to communicate using the routes of its neighbors. When a node joins the mesh network, Netsukuku automatically adapts and all other nodes come to know the fastest and most efficient routes to communicate with the newcomer. Each node has no more privileges or restrictions than the other nodes. The domain name system (DNS) is replaced by a decentralised and distributed system called ANDNA (Abnormal Netsukuku Domain Name Anarchy). The ANDNA database is included in the Netsukuku system, so each node includes such database that occupies at most 355 kilobytes of memory. Simplifying, ANDNA works as follows: to resolve a symbolic name the host applies a function Hash on its behalf. The Hash function returns an address that the host contacts asking for the resolution generated by the hash. The contacted node receives a request, searches in its ANDNA database for the address associated with the name and returns it to the applicant host. Recording works in a similar way: for example, let's suppose that the node X wants to register the address FreakNet.andna; X calculates the hash name and obtains the address 11.22.33.44 associated with node Y. The node X contacts Y asking to register 11.22.33.44 as its own. Y stores the request in its database and any request for resolution of 11.22.33.44 hash, will answer with the X's address. The protocol is a little more complex than this, as the system provides a public/private key to authenticate the hosts and prevent unauthorized changes to the ANDNA database. Furthermore, the protocol provides redundancy in the database to make the protocol resistant to failure and also provides for the migration of the database if the network topology changes. The protocol does not provide for the possibility of revoking a symbolic name; after a certain period of inactivity (currently 3 days) it is simply deleted from the database. The protocol also prevents a single host from recording an excessive number of symbolic names (at present 256 names) in order to prevent spammers from storing a high number of terms to perform cybersquatting.
Read more →
Social media use in education

Social media in education is the use of social media to enhance education. Social media are "a group of Internet-based applications...that allow the creation and exchange of user-generated content". It is also known as the read/write web. As time went on and technology evolved, social media has been an integral part of people's lives, including students, scholars, and teachers. However, social media are controversial because, in addition to providing new means of connection, critics claim that they damage self-esteem, shorten attention spans, and increase mental health issues. A 2016 dissertation presented surveys that focused on the impact of social media. It reported that 54.6% of students believed that social media affected their studies positively (38% agree, 16.6% strongly agree). About 40% disagreed, and 4.7% of students strongly disagreed. 53% of female students reported that social media negatively impacted their studies. Among male students, 40% agreed that social media had a negative impact on studies, while 59% disagreed. A 2023 article dives deep into the rewards system of the brain in response to social media. This study compares the social rewards system in our brain to those from social media. From ages 10-12, most are receiving a cell phone, social rewards in the brain start to feel more satisfying. Leading to adulthood, the effects of social rewards are less likely to feel reliant on feedback from peers. Equivalent to a more mature prefrontal cortex, this enables a better management of their emotional reaction to these social rewards, meaning a more balanced and controlled reaction. == History == A survey from Cambridge International of nearly 20,000 teachers and students (ages 12–19) from 100 countries found that 48% of students use a desktop computer in class, 42% uses phones, 33% use interactive whiteboards and 20% use tablets. Desktop computers are more used than tablets. Teachers were abandoning the "no phones at school" rule. A 2024 research survey through Common Sense Education reported 54% of age 8-12 and 69% of ages 13-18 social media is an extensive distraction from homework. === United States === The long-running technology boom accelerated after the millennium. As of 2018, 95% of US teenage students had access to a smartphone and 45% said they were online almost constantly. In the early days of social media, access to technology was a significant issue as many students did not own not compatible devices and school budgets were often insufficient to purchase devices for student use. Despite backlash, Missouri passed a law that prohibited teachers from communicating privately with students over social media in 2011. Supporters were concerned that online communication between underage students and faculty could lead to inappropriate relationships. Some schools adopted a "Bring Your Own Device" (BYOD) policy, allowing students to bring Internet-accessing devices, such as phones or tablets to class. During the pandemic, the federal government offered funds that allowed more schools to purchase devices. Over time, more students acquired phones with social media access. Personal devices increased student satisfaction, but reduced teachers' ability to control device use in their classrooms. A 2018 Pew Research study reported that 95% of teenagers had a phone and used social media consistently. === Canada === The Peel District School Board (PDSB) in Ontario accepted the use of social media in the classroom. In 2013, the PDSB introduced BYOD and unblocked many social media sites. That was later replaced by a policy that dealt specifically with social media. == Uses == === Classroom === In the classroom, social media offers a way to systematically distribute and gather information from students. Teachers can supply documents, and audio/video media to students for immediate or later use. One study on higher education reported that devices and social media: created opportunities for interaction provided occasions for collaboration sped up information access offered more ways to learn situated learning. Frustrations included anti-technology instructors, device challenges, and devices as a distraction. Social media in classrooms can have a negative effect. A Yale University publication reported that students who used laptops in class for non-academic reasons had poorer performance. Students spent most of their time on social media, shopping, and other personal activities. Social media has helped many educators mentor their students more effectively. === Outside of class === Social media offer a venue for video calls, stories, feeds, and game playing that can enhance the learning process. Teachers can utilize social media to communicate with their students. Social media can provide students with resources that they can utilize in essays, projects, and presentations. Students can easily access comments made by teachers and peers and offer feedback to teachers. Social media can offer students the opportunity to collaborate by sharing information without requiring face to face meetings. Social media can allow students to more easily connect with experts, to go beyond course materials. Instructors in a 2010 study reported that online technologies (social media) can help students become comfortable having discussions outside the classroom better than traditional means. Teachers may face some risk when using social media outside the classroom, without appropriate work rules. Studies explores how college students' engagement with social media platforms influences their communication preferences and habits, particularly in relation to using school email for academic purposes. === Professional development === Social media can aid professional development, as teachers become students, enhancing knowledge transfer, skill master, and collaboration. === Non-academic uses === Schools can use social media to make public announcements. Teachers and administrators can communicate other important information to parents and students and to receive feedback from them. Families can keep up with school events and policies. === Ecology education === The potential of using social media in ecological, nature and forest education include: virtual nature groups can help promote good habits in forest tourism and recreation (nature ethics), by entering general rules in the regulations by administrators, e.g. "DO NOT PICK UP PLANTS UNKNOWN TO US", which is to protects rare species from pointless picking. social media activity motivates people to learn about nature in the field, allows them to gain knowledge, dispels popular myths, enables contact with scientists and practitioners, promotes valuable literature, websites, and at the same time reveals distortions and substantive errors in popular news services. contact is not only virtual. Despite financial barriers and distance, Internet users organize nature conventions. Such meetings are an opportunity not only to make friends, but also to learn about nature together and have fun. the possibility of contact between scientists and nature lovers via Facebook has become a source of cooperation in species inventory, e.g. the online campaign of the NATRIX Herpetological Society, which consists not only of collecting reports of observations of the smooth snake by Internet users, but also of drawing attention to the biology and threats to this species. Social media has become a place where ecology education quickly reaches people of different ages and social statuses. The nature groups that have been created, in which nature lovers, biologists, foresters and scientists participate, can have a real impact on the state of knowledge and data collection through citizen science. == Apps and services == Social media can allow students to participate in their field by working with organizations outside the classroom. By offering easier access to peers outside the classroom, students can broaden their perspectives and find support resources. Social media aided learning outside of the classroom through collaboration and innovation. One specific study, "Exploring education-related use of social media," called this "audience connectors". Audience connectors bring students together while studying with WhatsApp and Facebook. This study reported that "60 percent [of students in the study] agreed that technology changes education for the better." While social media can promote a beneficial education platform, downsides exist. Students may become skilled at "lifting material from the internet" rather than enhancing their personal understanding. Another downside is student attention spans decline. A concern raised by the students of this study showed how many use spell-check as a crutch and will see a trend of points taken off when spell-check is not an option. Apps like X allowed teachers to make classroom accounts where students can learn about social media in a controlled context. Teachers can post assignments on th
Read more →
Cryptosystem

In cryptography, a cryptosystem is a suite of cryptographic algorithms needed to implement a particular security service, such as confidentiality (encryption). Typically, a cryptosystem consists of three algorithms: one for key generation, one for encryption, and one for decryption. The term cipher (sometimes cypher) is often used to refer to a pair of algorithms, one for encryption and one for decryption. Therefore, the term cryptosystem is most often used when the key generation algorithm is important. For this reason, the term cryptosystem is commonly used to refer to public key techniques; however both "cipher" and "cryptosystem" are used for symmetric key techniques. == Formal definition == Mathematically, a cryptosystem or encryption scheme can be defined as a tuple ( P , C , K , E , D ) {\displaystyle ({\mathcal {P}},{\mathcal {C}},{\mathcal {K}},{\mathcal {E}},{\mathcal {D}})} with the following properties. P {\displaystyle {\mathcal {P}}} is a set called the "plaintext space". Its elements are called plaintexts. C {\displaystyle {\mathcal {C}}} is a set called the "ciphertext space". Its elements are called ciphertexts. K {\displaystyle {\mathcal {K}}} is a set called the "key space". Its elements are called keys. E = { E k : k ∈ K } {\displaystyle {\mathcal {E}}=\{E_{k}:k\in {\mathcal {K}}\}} is a set of functions E k : P → C {\displaystyle E_{k}:{\mathcal {P}}\rightarrow {\mathcal {C}}} . Its elements are called "encryption functions". D = { D k : k ∈ K } {\displaystyle {\mathcal {D}}=\{D_{k}:k\in {\mathcal {K}}\}} is a set of functions D k : C → P {\displaystyle D_{k}:{\mathcal {C}}\rightarrow {\mathcal {P}}} . Its elements are called "decryption functions". For each e ∈ K {\displaystyle e\in {\mathcal {K}}} , there is d ∈ K {\displaystyle d\in {\mathcal {K}}} such that D d ( E e ( p ) ) = p {\displaystyle D_{d}(E_{e}(p))=p} for all p ∈ P {\displaystyle p\in {\mathcal {P}}} . Note; typically this definition is modified in order to distinguish an encryption scheme as being either a symmetric-key or public-key type of cryptosystem. == Examples == A classical example of a cryptosystem is the Caesar cipher. A more contemporary example is the RSA cryptosystem. Another example of a cryptosystem is the Advanced Encryption Standard (AES). AES is a widely used symmetric encryption algorithm that has become the standard for securing data in various applications. Paillier cryptosystem is another example used to preserve and maintain privacy and sensitive information. It is featured in electronic voting, electronic lotteries and electronic auctions.
Read more →
JAX (software)

JAX is a Python library for accelerator-oriented array computation and program transformation, designed for high-performance numerical computing and large-scale machine learning. It is developed by Google with contributions from Nvidia and other community contributors. It is described as bringing together a modified version of the automatic differentiation system autograd and OpenXLA's XLA (Accelerated Linear Algebra). It is designed to follow the structure and workflow of NumPy as closely as possible and works with various existing frameworks such as TensorFlow and PyTorch. The primary features of JAX are: Providing a unified NumPy-like interface to computations that run on CPU, GPU, or TPU, in local or distributed settings. Built-in Just-In-Time (JIT) compilation via OpenXLA, an open-source machine learning compiler ecosystem. Efficient evaluation of gradients via its automatic differentiation transformations. Automatic vectorization to efficiently map functions over arrays representing batches of inputs. == Libraries using Jax == Flax Equinox Optax
Read more →
Trace zero cryptography

First proposed by Gerhard Frey in 1998, trace zero cryptography refers to the use of trace zero varieties (TZV) for cryptographic purpose. Trace zero varieties are subgroups of the divisor class group on a low genus hyperelliptic curve defined over a finite field. These groups can be used to establish asymmetric cryptography using the discrete logarithm problem as cryptographic primitive. Trace zero varieties feature a better scalar multiplication performance than elliptic curves. This allows fast arithmetic in these groups, which can speed up the calculations with a factor 3 compared with elliptic curves and hence speed up the cryptosystem. Another advantage is that for groups of cryptographically relevant size, the order of the group can simply be calculated using the characteristic polynomial of the Frobenius endomorphism. This is not the case, for example, in elliptic curve cryptography when the group of points of an elliptic curve over a prime field is used for cryptographic purpose. However, to represent an element of the trace zero variety more bits are needed compared with elements of elliptic or hyperelliptic curves. Another disadvantage is the fact that it is possible to reduce the security of the TZV of 1/6th of the bit length using cover attack. == Mathematical background == A hyperelliptic curve C of genus g over a prime field F q {\displaystyle \mathbb {F} _{q}} where q = pn (p prime) of odd characteristic is defined as C : y 2 + h ( x ) y = f ( x ) , {\displaystyle C:~y^{2}+h(x)y=f(x),} where f monic, deg(f) = 2g + 1 and deg(h) ≤ g. The curve has at least one F q {\displaystyle \mathbb {F} _{q}} -rational Weierstraßpoint. The Jacobian variety J C ( F q n ) {\displaystyle J_{C}(\mathbb {F} _{q^{n}})} of C is for all finite extension F q n {\displaystyle \mathbb {F} _{q^{n}}} isomorphic to the ideal class group Cl ⁡ ( C / F q n ) {\displaystyle \operatorname {Cl} (C/\mathbb {F} _{q^{n}})} . With the Mumford's representation it is possible to represent the elements of J C ( F q n ) {\displaystyle J_{C}(\mathbb {F} _{q^{n}})} with a pair of polynomials [u, v], where u, v ∈ F q n [ x ] {\displaystyle \mathbb {F} _{q^{n}}[x]} . The Frobenius endomorphism σ is used on an element [u, v] of J C ( F q n ) {\displaystyle J_{C}(\mathbb {F} _{q^{n}})} to raise the power of each coefficient of that element to q: σ([u, v]) = [uq(x), vq(x)]. The characteristic polynomial of this endomorphism has the following form: χ ( T ) = T 2 g + a 1 T 2 g − 1 + ⋯ + a g T g + ⋯ + a 1 q g − 1 T + q g , {\displaystyle \chi (T)=T^{2g}+a_{1}T^{2g-1}+\cdots +a_{g}T^{g}+\cdots +a_{1}q^{g-1}T+q^{g},} where ai in Z {\displaystyle \mathbb {Z} } With the Hasse–Weil theorem it is possible to receive the group order of any extension field F q n {\displaystyle \mathbb {F} _{q^{n}}} by using the complex roots τi of χ(T): | J C ( F q n ) | = ∏ i = 1 2 g ( 1 − τ i n ) {\displaystyle |J_{C}(\mathbb {F} _{q^{n}})|=\prod _{i=1}^{2g}(1-\tau _{i}^{n})} Let D be an element of the J C ( F q n ) {\displaystyle J_{C}(\mathbb {F} _{q^{n}})} of C, then it is possible to define an endomorphism of J C ( F q n ) {\displaystyle J_{C}(\mathbb {F} _{q^{n}})} , the so-called trace of D: Tr ⁡ ( D ) = ∑ i = 0 n − 1 σ i ( D ) = D + σ ( D ) + ⋯ + σ n − 1 ( D ) {\displaystyle \operatorname {Tr} (D)=\sum _{i=0}^{n-1}\sigma ^{i}(D)=D+\sigma (D)+\cdots +\sigma ^{n-1}(D)} Based on this endomorphism one can reduce the Jacobian variety to a subgroup G with the property, that every element is of trace zero: G = { D ∈ J C ( F q n ) | Tr ( D ) = 0 } , ( 0 neutral element in J C ( F q n ) {\displaystyle G=\{D\in J_{C}(\mathbb {F} _{q^{n}})~|~{\text{Tr}}(D)={\textbf {0}}\},~~~({\textbf {0}}{\text{ neutral element in }}J_{C}(\mathbb {F} _{q^{n}})} G is the kernel of the trace endomorphism and thus G is a group, the so-called trace zero (sub)variety (TZV) of J C ( F q n ) {\displaystyle J_{C}(\mathbb {F} _{q^{n}})} . The intersection of G and J C ( F q ) {\displaystyle J_{C}(\mathbb {F} _{q})} is produced by the n-torsion elements of J C ( F q ) {\displaystyle J_{C}(\mathbb {F} _{q})} . If the greatest common divisor gcd ( n , | J C ( F q ) | ) = 1 {\displaystyle \gcd(n,|J_{C}(\mathbb {F} _{q})|)=1} the intersection is empty and one can compute the group order of G: | G | = | J C ( F q n ) | | J C ( F q ) | = ∏ i = 1 2 g ( 1 − τ i n ) ∏ i = 1 2 g ( 1 − τ i ) {\displaystyle |G|={\dfrac {|J_{C}(\mathbb {F} _{q^{n}})|}{|J_{C}(\mathbb {F} _{q})|}}={\dfrac {\prod _{i=1}^{2g}(1-\tau _{i}^{n})}{\prod _{i=1}^{2g}(1-\tau _{i})}}} The actual group used in cryptographic applications is a subgroup G0 of G of a large prime order l. This group may be G itself. There exist three different cases of cryptographical relevance for TZV: g = 1, n = 3 g = 1, n = 5 g = 2, n = 3 == Arithmetic == The arithmetic used in the TZV group G0 based on the arithmetic for the whole group J C ( F q n ) {\displaystyle J_{C}(\mathbb {F} _{q^{n}})} , But it is possible to use the Frobenius endomorphism σ to speed up the scalar multiplication. This can be archived if G0 is generated by D of order l then σ(D) = sD, for some integers s. For the given cases of TZV s can be computed as follows, where ai come from the characteristic polynomial of the Frobenius endomorphism : For g = 1, n = 3: s = q − 1 1 − a 1 mod ℓ {\displaystyle s={\dfrac {q-1}{1-a_{1}}}{\bmod {\ell }}} For g = 1, n = 5: s = q 2 − q − a 1 2 q + a 1 q + 1 q − 2 a 1 q + a 1 3 − a 1 2 + a 1 − 1 mod ℓ {\displaystyle s={\dfrac {q^{2}-q-a_{1}^{2}q+a_{1}q+1}{q-2a_{1}q+a_{1}^{3}-a_{1}^{2}+a_{1}-1}}{\bmod {\ell }}} For g = 2, n = 3: s = − q 2 − a 2 + a 1 a 1 q − a 2 + 1 mod ℓ {\displaystyle s=-{\dfrac {q^{2}-a_{2}+a_{1}}{a_{1}q-a_{2}+1}}{\bmod {\ell }}} Knowing this, it is possible to replace any scalar multiplication mD (|m| ≤ l/2) with: m 0 D + m 1 σ ( D ) + ⋯ + m n − 1 σ n − 1 ( D ) , where m i = O ( ℓ 1 / ( n − 1 ) ) = O ( q g ) {\displaystyle m_{0}D+m_{1}\sigma (D)+\cdots +m_{n-1}\sigma ^{n-1}(D),~~~~{\text{where }}m_{i}=O(\ell ^{1/(n-1)})=O(q^{g})} With this trick the multiple scalar product can be reduced to about 1/(n − 1)th of doublings necessary for calculating mD, if the implied constants are small enough. == Security == The security of cryptographic systems based on trace zero subvarieties according to the results of the papers comparable to the security of hyper-elliptic curves of low genus g' over F p ′ {\displaystyle \mathbb {F} _{p'}} , where p' ~ (n − 1)(g/g' ) for |G| ~128 bits. For the cases where n = 3, g = 2 and n = 5, g = 1 it is possible to reduce the security for at most 6 bits, where |G| ~ 2256, because one can not be sure that G is contained in a Jacobian of a curve of genus 6. The security of curves of genus 4 for similar fields are far less secure. == Cover attack on a trace zero crypto-system == The attack published in shows, that the DLP in trace zero groups of genus 2 over finite fields of characteristic diverse than 2 or 3 and a field extension of degree 3 can be transformed into a DLP in a class group of degree 0 with genus of at most 6 over the base field. In this new class group the DLP can be attacked with the index calculus methods. This leads to a reduction of the bit length 1/6th.
Read more →
Format-transforming encryption

In cryptography, format-transforming encryption (FTE) refers to encryption where the format of the input plaintext and output ciphertext are configurable. Descriptions of formats can vary, but are typically compact set descriptors, such as a regular expression. Format-transforming encryption is closely related to, and a generalization of, format-preserving encryption. == Applications of FTE == === Restricted fields or formats === Similar to format-preserving encryption, FTE can be used to control the format of ciphertexts. The canonical example is a credit card number, such as 1234567812345670 (16 bytes long, digits only). However, FTE does not enforce that the input format must be the same as the output format. === Censorship circumvention === FTE is used by the Tor Project to circumvent deep packet inspection by pretending to be some other protocols. The implementation is fteproxy; it was written by the authors who came up with the FTE concept.
Read more →
Backdoor (computing)

A backdoor is a typically covert method of bypassing normal authentication or encryption in a computer, product, embedded device (e.g. a home router), or its embodiment (e.g. part of a cryptosystem, algorithm, chipset, or even a "homunculus computer"—a tiny computer-within-a-computer such as that found in Intel's AMT technology). Backdoors are most often used for securing remote access to a computer, or obtaining access to plaintext in cryptosystems. From there it may be used to gain access to privileged information like passwords, corrupt or delete data on hard drives, or transfer information within compromised networks. In the United States, the 1994 Communications Assistance for Law Enforcement Act forces internet providers to provide backdoors for government authorities. In 2024, the U.S. government realized that China had been tapping communications in the U.S. using that infrastructure for months, or perhaps longer; China recorded presidential candidate campaign office phone calls—including employees of the then-vice president of the nation, and of the candidates themselves. A backdoor may take the form of a hidden part of a program, a separate program (e.g. Back Orifice may subvert the system through a rootkit), code in the firmware of the hardware, or parts of an operating system such as Windows, for example, device drivers. Trojan horses can be used to create vulnerabilities in a device. A Trojan horse may appear to be an entirely legitimate program, but when executed, it triggers an activity that may install a backdoor. Although some are secretly installed, other backdoors are deliberate and widely known. These kinds of backdoors have "legitimate" uses such as providing the manufacturer with a way to restore user passwords. Many systems that store information within the cloud fail to create accurate security measures. If many systems are connected within the cloud, hackers can gain access to all other platforms through the most vulnerable system. Default passwords (or other default credentials) can function as backdoors if they are not changed by the user. Some debugging features can also act as backdoors if they are not removed in the release version. In 1993, the United States government attempted to deploy an encryption system, the Clipper chip, with an explicit backdoor for law enforcement and national security access. The chip was unsuccessful. Recent proposals to counter backdoors include creating a database of backdoors' triggers and then using neural networks to detect them. == Overview == The threat of backdoors surfaced when multiuser and networked operating systems became widely adopted. Petersen and Turn discussed computer subversion in a paper published in the proceedings of the 1967 AFIPS Conference. They noted a class of active infiltration attacks that use "trapdoor" entry points into the system to bypass security facilities and permit direct access to data. The use of the word trapdoor here clearly coincides with more recent definitions of a backdoor. However, since the advent of public key cryptography the term trapdoor has acquired a different meaning (see: Trapdoor function), and thus the term "backdoor" is now preferred, only after the term trapdoor went out of use. More generally, such security breaches were discussed at length in a RAND Corporation task force report published under DARPA sponsorship by J.P. Anderson and D.J. Edwards in 1970. While initially targeting the computer vision domain, backdoor attacks have expanded to encompass various other domains, including text, audio, ML-based computer-aided design, and ML-based wireless signal classification. Additionally, vulnerabilities in backdoors have been demonstrated in deep generative models, reinforcement learning (e.g., AI GO), and deep graph models. These broad-ranging potential risks have prompted concerns from national security agencies regarding their potentially disastrous consequences. A backdoor in a login system might take the form of a hard coded user and password combination which gives access to the system. An example of this sort of backdoor was used as a plot device in the 1983 film WarGames, in which the architect of the "WOPR" computer system had inserted a hardcoded password-less account which gave the user access to the system, and to undocumented parts of the system (in particular, a video game-like simulation mode and direct interaction with the artificial intelligence). Although the number of backdoors in systems using proprietary software (software whose source code is not publicly available) is not widely credited, they are nevertheless frequently exposed. Programmers have even succeeded in secretly installing large amounts of benign code as Easter eggs in programs, although such cases may involve official forbearance, if not actual permission. == Examples == === Worms === Many computer worms, such as Sobig and Mydoom, install a backdoor on the affected computer (generally a PC on broadband running Microsoft Windows and Microsoft Outlook). Such backdoors appear to be installed so that spammers can send junk e-mail from the infected machines. Others, such as the Sony/BMG rootkit, placed secretly on millions of music CDs through late 2005, are intended as DRM measures—and, in that case, as data-gathering agents, since both surreptitious programs they installed routinely contacted central servers. A sophisticated attempt to plant a backdoor in the Linux kernel, exposed in November 2003, added a small and subtle code change by subverting the revision control system. In this case, a two-line change appeared to check root access permissions of a caller to the sys_wait4 function, but because it used assignment = instead of equality checking ==, it actually granted permissions to the system. This difference is easily overlooked, and could even be interpreted as an accidental typographical error, rather than an intentional attack. In January 2014, a backdoor was discovered in certain Samsung Android products, like the Galaxy devices. The Samsung proprietary Android versions are fitted with a backdoor that provides remote access to the data stored on the device. In particular, the Samsung Android software that is in charge of handling the communications with the modem, using the Samsung IPC protocol, implements a class of requests known as remote file server (RFS) commands, that allows the backdoor operator to perform via modem remote I/O operations on the device hard disk or other storage. As the modem is running Samsung proprietary Android software, it is likely that it offers over-the-air remote control that could then be used to issue the RFS commands and thus to access the file system on the device. === Object code backdoors === Harder to detect backdoors involve modifying object code, rather than source code—object code is much harder to inspect, as it is designed to be machine-readable, not human-readable. These backdoors can be inserted either directly in the on-disk object code, or inserted at some point during compilation, assembly linking, or loading—in the latter case the backdoor never appears on disk, only in memory. Object code backdoors are difficult to detect by inspection of the object code, but are easily detected by simply checking for changes (differences), notably in length or in checksum, and in some cases can be detected or analyzed by disassembling the object code. Further, object code backdoors can be removed (assuming source code is available) by simply recompiling from source on a trusted system. Thus for such backdoors to avoid detection, all extant copies of a binary must be subverted, and any validation checksums must also be compromised, and source must be unavailable, to prevent recompilation. Alternatively, these other tools (length checks, diff, checksumming, disassemblers) can themselves be compromised to conceal the backdoor, for example detecting that the subverted binary is being checksummed and returning the expected value, not the actual value. To conceal these further subversions, the tools must also conceal the changes in themselves—for example, a subverted checksummer must also detect if it is checksumming itself (or other subverted tools) and return false values. This leads to extensive changes in the system and tools being needed to conceal a single change. As object code can be regenerated by recompiling (reassembling, relinking) the original source code, making a persistent object code backdoor (without modifying source code) requires subverting the compiler itself—so that when it detects that it is compiling the program under attack it inserts the backdoor—or alternatively the assembler, linker, or loader. As this requires subverting the compiler, this in turn can be fixed by recompiling the compiler, removing the backdoor insertion code. This defense can in turn be subverted by putting a source meta-backdoor in the compiler, so that when it detects that it is compiling itself
Read more →
Self-supervised learning

Self-supervised learning (SSL) is a paradigm in machine learning where a model is trained on a task using the data itself to generate supervisory signals, rather than relying on externally-provided labels. In the context of neural networks, self-supervised learning aims to leverage inherent structures or relationships within the input data to create meaningful training signals. SSL tasks are designed so that solving them requires capturing essential features or relationships in the data. The input data is typically augmented or transformed in a way that creates pairs of related samples, where one sample serves as the input, and the other is used to formulate the supervisory signal. This augmentation can involve introducing noise, cropping, rotation, or other transformations. Self-supervised learning more closely imitates the way humans learn to classify objects. During SSL, the model learns in two steps. First, the task is solved based on an auxiliary or pretext classification task using pseudo-labels, which help to initialize the model parameters. Next, the actual task is performed with supervised or unsupervised learning. Self-supervised learning has produced promising results in recent years, and has found practical application in fields such as audio processing, and is being used by Facebook and others for speech recognition. == Pseudo-labels == Pseudo-labels are automatically generated labels that a model assigns to unlabeled data based on its own predictions. They are widely used in self-supervised and semi-supervised learning, where ground-truth annotations are limited or unavailable. By treating predicted labels as surrogate ground truth, learning algorithms can make use of large quantities of unlabeled data in the training process. Pseudo-labeling also plays an important role in systems that must adapt to concept drift, where the statistical properties of the data change over time. In these scenarios, the model may detect that an incoming instance deviates from previously learned behavior. The system then generates a classification result for that instance, and this predicted class is used as a pseudo-label for updating or retraining model components that are becoming outdated. This approach enables continuous adaptation in dynamic environments without requiring manual annotation. In many adaptive learning pipelines, pseudo-labels are chosen when the classifier produces sufficiently confident predictions, reducing the risk of propagating errors. These pseudo-labeled instances are then incorporated into training to refresh or evolve the model's understanding of emerging data patterns, particularly when existing components show signs of “aging” due to drift or distributional shifts. This strategy reduces reliance on manual labeling while helping maintain long-term model performance. == Types == === Autoassociative self-supervised learning === Autoassociative self-supervised learning is a specific category of self-supervised learning where a neural network is trained to reproduce or reconstruct its own input data. In other words, the model is tasked with learning a representation of the data that captures its essential features or structure, allowing it to regenerate the original input. The term "autoassociative" comes from the fact that the model is essentially associating the input data with itself. This is often achieved using autoencoders, which are a type of neural network architecture used for representation learning. Autoencoders consist of an encoder network that maps the input data to a lower-dimensional representation (latent space), and a decoder network that reconstructs the input from this representation. The training process involves presenting the model with input data and requiring it to reconstruct the same data as closely as possible. The loss function used during training typically penalizes the difference between the original input and the reconstructed output (e.g. mean squared error). By minimizing this reconstruction error, the autoencoder learns a meaningful representation of the data in its latent space. === Contrastive self-supervised learning === For a binary classification task, training data can be divided into positive examples and negative examples. Positive examples are those that match the target. For example, if training a classifier to identify birds, the positive training data would include images that contain birds. Negative examples would be images that do not. Contrastive self-supervised learning uses both positive and negative examples. The loss function in contrastive learning is used to minimize the distance between positive sample pairs, while maximizing the distance between negative sample pairs. An early example uses a pair of 1-dimensional convolutional neural networks to process a pair of images and maximize their agreement. Contrastive Language-Image Pre-training (CLIP) allows joint pretraining of a text encoder and an image encoder, such that a matching image-text pair have image encoding vector and text encoding vector that span a small angle (having a large cosine similarity). InfoNCE (Noise-Contrastive Estimation) is a method to optimize two models jointly, based on Noise Contrastive Estimation (NCE). Given a set X = { x 1 , … x N } {\displaystyle X=\left\{x_{1},\ldots x_{N}\right\}} of N {\displaystyle N} random samples containing one positive sample from p ( x t + k ∣ c t ) {\displaystyle p\left(x_{t+k}\mid c_{t}\right)} and N − 1 {\displaystyle N-1} negative samples from the 'proposal' distribution p ( x t + k ) {\displaystyle p\left(x_{t+k}\right)} , it minimizes the following loss function: L N = − E X [ log ⁡ f k ( x t + k , c t ) ∑ x j ∈ X f k ( x j , c t ) ] {\displaystyle {\mathcal {L}}_{\mathrm {N} }=-\mathbb {E} _{X}\left[\log {\frac {f_{k}\left(x_{t+k},c_{t}\right)}{\sum _{x_{j}\in X}f_{k}\left(x_{j},c_{t}\right)}}\right]} === Non-contrastive self-supervised learning === Non-contrastive self-supervised learning (NCSSL) uses only positive examples. Counterintuitively, NCSSL converges on a useful local minimum rather than reaching a trivial solution, with zero loss. For the example of binary classification, it would trivially learn to classify each example as positive. Effective NCSSL requires an extra predictor on the online side that does not back-propagate on the target side. === Joint-Embedding and Predictive Architectures === A major class of self-supervised learning moves beyond contrastive pairs, instead maximizing the agreement between views while preventing collapse through statistical constraints. Rooted in Deep Canonical Correlation Analysis (Deep CCA), this approach includes Joint-Embedding Architectures (JEA) like Barlow Twins and VICReg, which enforce covariance constraints to learn invariant representations without negative sampling. Deep Latent Variable Path Modelling (DLVPM) generalizes this to multimodal systems, using path models to enforce correlation and orthogonality across diverse data types. In 2022 Yann LeCun introduced Joint-Embedding Predictive Architectures (JEPA) as a step towards decision making, reasoning, and autonomous human intelligence in machines, including self-improvement through autonomous learning. Founded in representation learning, LeCun included the concept of a “world model” in JEPA which aims to enable machines to replicate human intellect by providing machines with a concept for the world in which they exist. Unlike autoencoders, JEPAs operate entirely in latent space, avoiding pixel-level noise to focus on semantic structure. Rather than just learning invariance, JEPAs learn by predicting masked latent representations from visible context. JEPA has been applied to domains such as image analysis, audio processing, and motion in images and video. == Comparison with other forms of machine learning == SSL belongs to supervised learning methods insofar as the goal is to generate a classified output from the input. At the same time, however, it does not require the explicit use of labeled input-output pairs. Instead, correlations, metadata embedded in the data, or domain knowledge present in the input are implicitly and autonomously extracted from the data. These supervisory signals, extracted from the data, can then be used for training. SSL is similar to unsupervised learning in that it does not require labels in the sample data. Unlike unsupervised learning, however, learning is not done using inherent data structures. Semi-supervised learning combines supervised and unsupervised learning, requiring only a small portion of the learning data be labeled. In transfer learning, a model designed for one task is reused on a different task. Training an autoencoder intrinsically constitutes a self-supervised process, because the output pattern needs to become an optimal reconstruction of the input pattern itself. However, in current jargon, the term 'self-supervised' often refers to tasks based on a pretext-task training setup
Read more →
Netsukuku

Netsukuku is an experimental peer-to-peer routing system, developed by the FreakNet MediaLab in 2005, created to build up a distributed network, anonymous and censorship-free, fully independent but not necessarily separated from the Internet, without the support of any server, Internet service provider and no central authority. Netsukuku is designed to handle up to 2128 nodes without any servers or central systems, with minimal CPU and memory resources. This mesh network can be built using existing network infrastructure components such as Wi-Fi. The project has been in slow development since 2005, never abandoning a beta state. It has also never been tested on large scale. == Operation == As of December 2011, the latest theoretical work on Netsukuku could be found in the author's master thesis Scalable Mesh Networks and the Address Space Balancing problem. The following description takes into account only the basic concepts of the theory. Netsukuku uses a custom routing protocol called QSPN (Quantum Shortest Path Netsukuku) that strives to be efficient and not taxing on the computational capabilities of each node. The current version of the protocol is QSPNv2. It adopts a hierarchical structure. 256 nodes are grouped inside a gnode (group node), 256 gnodes are grouped in a single ggnode (group of group nodes), 256 ggnodes are grouped in a single gggnode, and so on. This offers a set of advantages main documentation. The protocol relies on the fact that the nodes are not mobile and that the network structure does not change quickly, as several minutes may be required before a change in the network is propagated. However, a node that joins the network is immediately able to communicate using the routes of its neighbors. When a node joins the mesh network, Netsukuku automatically adapts and all other nodes come to know the fastest and most efficient routes to communicate with the newcomer. Each node has no more privileges or restrictions than the other nodes. The domain name system (DNS) is replaced by a decentralised and distributed system called ANDNA (Abnormal Netsukuku Domain Name Anarchy). The ANDNA database is included in the Netsukuku system, so each node includes such database that occupies at most 355 kilobytes of memory. Simplifying, ANDNA works as follows: to resolve a symbolic name the host applies a function Hash on its behalf. The Hash function returns an address that the host contacts asking for the resolution generated by the hash. The contacted node receives a request, searches in its ANDNA database for the address associated with the name and returns it to the applicant host. Recording works in a similar way: for example, let's suppose that the node X wants to register the address FreakNet.andna; X calculates the hash name and obtains the address 11.22.33.44 associated with node Y. The node X contacts Y asking to register 11.22.33.44 as its own. Y stores the request in its database and any request for resolution of 11.22.33.44 hash, will answer with the X's address. The protocol is a little more complex than this, as the system provides a public/private key to authenticate the hosts and prevent unauthorized changes to the ANDNA database. Furthermore, the protocol provides redundancy in the database to make the protocol resistant to failure and also provides for the migration of the database if the network topology changes. The protocol does not provide for the possibility of revoking a symbolic name; after a certain period of inactivity (currently 3 days) it is simply deleted from the database. The protocol also prevents a single host from recording an excessive number of symbolic names (at present 256 names) in order to prevent spammers from storing a high number of terms to perform cybersquatting.
Read more →
Data definition specification

In computing, a data definition specification (DDS) is a guideline to ensure comprehensive and consistent data definition. It represents the attributes required to quantify data definition. A comprehensive data definition specification encompasses enterprise data, the hierarchy of data management, prescribed guidance enforcement and criteria to determine compliance. == Overview == A data definition specification may be developed for any organization or specialized field, improving the quality of its products through consistency and transparency. It eliminates redundancy (since all contributing areas are referencing the same specification) and provides standardization and degrees of compliance, making it easier and more efficient to create, modify, verify, analyze and share information across the enterprise. To understand how a data definition specification works in an enterprise, we must look at the elements of a DDS. Writing data definitions, defining business terms (or rules) in the context of a particular environment, provides structure for an organization's data architecture. In developing these definitions, the words used must be traceable to clearly defined data. A data definition specification may be used in the following activities: Business intelligence Business process modeling Business rules management Data analysis and modeling Information architecture Metadata modeling Data mastering Report generation == Criteria == A data definition specification requires data definitions to be: Atomic – singular, describing only one concept. Commonly used and ambiguous terms should be defined. While a term refers to one concept, several words may be used in a term: File – A concept identifiable with one word File extension – A concept identifiable with more than one word Traceable – Mapped to a specific data element. In business, a term may be traced to an entity (for example, a customer) or an attribute (such as a customer's name). A term may be a value in a data set (such as gender), or designate the data set itself. Traceability indicates relationships in the data hierarchy. Consistent - Used in a standard syntax; if used in a specific context, the context is noted Accurate - Precise, correct and unambiguous, stating what the term is and is not Clear - Readily understood by the reader Complete - With the term, its description and contextual references Concise - To avoid circular references == Applications == === Enterprise data === A data definition specification was produced by the Open Mobile Alliance to document charging data. The document, the centralized catalog of data elements defined for interfaces, specifies the mapping of these data elements to protocol fields in the interfaces. Created for the exchange of financial data, Market Data Definition Language (MDDL) is an XML specification designed to enable the interchange of information necessary to account, to analyze, and to trade financial instruments of the world's markets. It defines an XML-based interchange format and common data dictionary on the fields needed to describe: (1) financial instruments, (2) corporate events affecting value and tradability, and (3) market-related, economic and industrial indicators. The principal function of MDDL is to allow entities to exchange market data by standardizing formats and definitions. MDDL provides a common format for market data so that it can be efficiently passed from one processing system to another and provides a common understanding of market data content by standardizing terminology and by normalizing the relationships of various data elements to one another ... From the user perspective, the goal of MDDL is to enable users to integrate data from multiple sources by standardizing both the input feeds used for data warehousing (i.e., define what's being provided by vendors) and the output methods by which client applications request the data (i.e., ensure compatibility on how to get data in and out of applications)." === Clinical submissions === The Clinical Data Interchange Standards Consortium, a global, multidisciplinary, non-profit organization, has established standards to support the acquisition, exchange, submission and archiving of clinical research data and metadata. CDISC standards are vendor-neutral, platform-independent and freely available from the CDISC website. The Case Report Tabulation Data Definition Specification (define.xml) draft version 2.0, the oldest data definition specification, is part of the evolution from the 1999 FDA electronic submission (eSub) guidance and electronic Common Technical Document (eCTD) documents specifying that a document describing the content and structure of included data be included in a submission. Define.xml was developed to automate the review process by generating a machine-readable data-definition document. Define.xml has standardized submissions to the Food and Drug Administration, reducing review times from over two years to several months. === Archival data === A data definition specification is the foundation of metadata for scientific data archiving. The Metadata Encoding and Transmission Standard (METS) uses one principle of a DDS: consistent use of key terms to catalog digital objects for global use. The METS schema is a flexible mechanism for encoding descriptive, administrative and structural metadata for a digital library object and expressing complex links between metadata, and can provide a useful standard for the exchange of digital-library objects between repositories. A similar effort is underway to preserve complex data associated with video-game archiving. Preserving Virtual Worlds attempted to address archival-format deficiencies, citing the lack of suitable documentation for interactive fiction and games at the bit level: specifically, the absence of "representation information" needed to map raw bits into higher-level data constructs. Preserving Virtual Worlds 2 is a research project expanding on initial efforts in this field.
Read more →
Experimental SAGE Subsector

The Experimental Semi-Automatic Ground Environment (SAGE) Sector (ESS, Experimental SAGE Subsector until planned Sectors/Subsectors were renamed NORAD Regions, Divisions, and Sectors) was a prototype Cold War Air Defense Sector for developing the Semi Automatic Ground Environment. The Lincoln Laboratory control center in a new building was at Lexington, Massachusetts. == ESS Computer System == The network's Direction Center was completed in a new 1954 building (Building F, 42°27′37″N 071°16′04″W) with prototype peripherals and a single IBM XD-1 computer, a successor to Lincoln Lab's Whirlwind I computer (WWI). In 1955, Air Force personnel began IBM training at the Kingston, New York, prototype facility, and the "4620th Air Defense Wing (experimental SAGE) was established at Lincoln Laboratory"—its "primary mission was computer programming". ESS had a capacity of 48 tracks and used a pre-SAGE ground environment in a "prototype intercept monitor room [at] MIT's Barta building" with "track situation displays, which geographically showed Air Defense Identification Zone lines and antiaircraft circles [and] each console also had a 5-inch CRT for digital information display. Audible alert signals were used, with a different signal for each symbol on a situation display." == Radar stations == Initial service test models of the Burroughs AN/FST-2 Coordinate Data Transmitting Set were placed with radars at South Truro and West Bath, Maine; followed by Texas Tower#2 (TT2) in the Atlantic Ocean, which provided a "triangular pattern with overlap" radar coverage (TT2 later had a connection from the XD-1 via the GE G/A Data Link Output Subsystem through North Truro Air Force Station.) By August 1955, 13 radar stations were networked by the subsector, e.g.: Chatham Clinton, Massachusetts with gap-filler radar Great Boars Head Halibut Point Killingly, Connecticut (41.865734°N 71.820958°W / 41.865734; -71.820958).with gap-filler radar Rockport Air Force Station Scituate, Massachusetts South Truro West Bath, Maine (43°54′7″N 69°50′43″W) with AN/FPS-31 on Jug Handle Hill: ("Lincoln Laboratories experimental radar station") Required by 21 November 1955 were 44 consoles: 38 for the operations floor, 3 on the computer floor for display maintenance, and 3 near the maintenance console (program checkout). WWI was connected to the Experimental SAGE Subsector to verify crosstelling (collateral communication) with the ESS DC, and WWI was also used for a Ground-to-Air (G/A) experiment using a transmitter of the GE G/A Data Link Output Subsystem on Prospect Hill, Waltham, MA sending data to simulated airborne equipment at Lexington. Transmissions from the WWI SAGE Evaluation (WISE) computer system to XD-1 and back were without error by December 1955 when operational software specifications were frozen. Operating procedures for the ESS external sites were complete in March 1956, and == System Operation Testing == From November 15, 1955, to November 7, 1956, three System Operation Tests were conducted which used voice "Ground-to-Air" communication from the Barta control room to aircraft outfitted with SAGE receivers (F-86 interceptors modified to F-86L models in "Project FOLLOW-ON".) Test teams included employees of Bell Telephone Laboratories, Western Electric-ADES, IBM, the RAND Corporation, and Lincoln Labs' Division 6, Division 3, & Division 2 (Division 6 had been created for ESS support.) The North Truro P-10 AN/FST-2 was moved to Almaden Air Force Station (M-96)c. 1957-8 and on August 7, 1958, control of an airborne BOMARC missile that had malfunctioned transferred from the "Experimental SAGE Sector" to a Westinghouse AN/GPA-35 Ground Environment system and the missile crashed into the Atlantic Ocean. By December 31, 1958, ADC Manual 55-28 described the Model 3 SAGE System. == 1959 Experimental Testing == "To prove out the revised SAGE computer program" for Automatic Targeting and Battery Evaluation and ADDC-AADCP crosstelling, a "SAGE/Missile Master" test was conducted beginning in September 1959 with communications between the ESS XD-1 and Martin AN/FSG-1 Antiaircraft Defense System equipment at Fort Banks planned for the CONAD Joint Control Center at Fort Heath—a "SAGE ATABE Simulation Study" (SASS) was also completed 1959–60 by MITRE Corporation.
Read more →
Grokking (machine learning)

In machine learning, grokking, or delayed generalization, is a phenomenon observed in some settings where a model abruptly transitions from overfitting (performing well only on training data) to generalizing (performing well on both training and test data), after many training iterations with little or no improvement on the held-out data. This contrasts with what is typically observed in machine learning, where generalization occurs gradually alongside improved performance on training data. == Origin == Grokking was introduced by OpenAI researcher Alethea Power and colleagues in the January 2022 paper "Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets". It is derived from the word grok coined by Robert Heinlein in his novel Stranger in a Strange Land. In ML research, "grokking" is not used as a synonym for "generalization"; rather, it names a sometimes-observed delayed‑generalization training phenomenon in which training and held‑out performance do not improve in tandem, and in which held‑out performance rises abruptly later. Authors also analyze the "grokking time", the epoch or step at which this transition occurs in those scenarios. == Interpretations == Grokking can be understood as a phase transition during the training process. In particular, recent work has shown that grokking may be due to a complexity phase transition in the model during training. While grokking has been thought of as largely a phenomenon of relatively shallow models, grokking has been observed in deep neural networks and non-neural models and is the subject of active research. One potential explanation is that the weight decay (a component of the loss function that penalizes higher values of the neural network parameters, also called regularization) slightly favors the general solution that involves lower weight values, but that is also harder to find. According to Neel Nanda, the process of learning the general solution may be gradual, even though the transition to the general solution occurs more suddenly later. Recent theories have hypothesized that grokking occurs when neural networks transition from a "lazy training" regime where the weights do not deviate far from initialization, to a "rich" regime where weights abruptly begin to move in task-relevant directions. Follow-up empirical and theoretical work has accumulated evidence in support of this perspective, and it offers a unifying view of earlier work as the transition from lazy to rich training dynamics is known to arise from properties of adaptive optimizers, weight decay, initial parameter weight norm, and more. This perspective is complementary to a unifying "pattern learning speeds" framework that links grokking and double descent; within this view, delayed generalization can arise across training time ("epoch‑wise") or across model size ("model‑wise"), and the authors report "model‑wise grokking".
Read more →
Data Management Association

The Data Management Association (DAMA), formerly known as the Data Administration Management Association, is a global not-for-profit organization which aims to advance concepts and practices about information management and data management. It describes itself as vendor-independent, all-volunteer organization, and has a membership consisting of technical and business professionals. Its international branch is called DAMA International (or DAMA-I), and DAMA also has various continental and national branches around the world. == History == The Data Management Association International was founded in 1980 in Los Angeles. Other early chapters were: San Francisco, Portland, Seattle, Minneapolis, New York, and Washington D.C. == Data Management Body of Knowledge == DAMA has published the Data Management Body of Knowledge (DMBOK), which contains suggestions on best practices and suggestions of a common vernacular for enterprise data management. The first edition (DAMA-DMBOK) was published on 2009 November 1, the second edition (DAMA-DMBOK2) was published on 2017 July 1., and the Revised second edition (DAMA-DMBOK2 rev.2) was published on 2019 March 19. DMBOK has been described by the authors as being an "equivalent" to the Project Management Body of Knowledge (PMBOK) and Business Analysis Body of Knowledge (BABOK). It encompasses topics such as data architecture, security, quality, modelling, governance, big data, data science, and more. DMBOK also includes the DAMA Data Wheel, an infographic which represents core data management practices. The center of the infographic is data governance, and the surrounding segments each represent a different aspect of data management: Data architecture Data modeling and design Data storage and operations Data security Data integration and interoperability Document management Content management Master data management Reference data and master data Data warehousing Metadata management Data quality Business intelligence Data science == Professional Accreditation == DAMA also provides a professional data management certification for individuals known as a Certified Data Management Professional (CDMP), which is based on the DMBOK as a study reference. There are four levels of certification based on career experience and exam results. The highest level, Fellow, requires 25 years of experience and nomination by DAMA members. It is an example of one of many competing certifications for data management professionals.
Read more →
Social network game

A social network game (sometimes simply referred to as a social media game, social gaming, or online social game) is a type of online game that is played through social networks or social media. They typically feature gamification systems with multiplayer gameplay mechanics. Social network games were originally implemented as browser games. As mobile gaming took off, the games moved to mobile as well. While they share many aspects of traditional video games, social network games often employ additional ones that make them distinct. Traditionally they are oriented to be social games and casual games. The first cross-platform "Facebook-to-Mobile" social network game was developed in 2011 by a Finnish company Star Arcade. Social network games are amongst the most popular games played in the world, with several products with tens of millions of players. (Lil) Green Patch, Happy Farm, and Mob Wars were some of the first successful games of this genre. FarmVille, Mafia Wars, Kantai Collection, and The Sims Social are more recent examples of popular social network game. Major companies that made or published social network games include Zynga, Wooga and Bigpoint Games. == Demographics == As of 2010, it was reported that 55 percent of the social network gaming demographic in the United States consisted of women while in the United Kingdom, women made up nearly 60 percent of the demographic. In addition, most social gamers were around the 30 to 59 age range, with the average social gamer being 43 years old. Social gaming may appeal more to the older demographic because it is free, easier to advance through in a short period, does not involve as much violence as traditional video games, and is easier to grasp. Other games target certain demographics that use social media, such as Pot Farm creating a community by involving elements of cannabis subculture in its gameplay. == Technology and platforms == A social network video game is a client-server application. The client in the web era was implemented with a mix of web technologies like Flash, HTML5, PHP and JavaScript. When mobile games moved to mobile, social game front ends were developed using mobile platform technologies like Java, Objective-C, Swift and C++. The back end was a mix of programming languages and systems, including PHP, Ruby, C++ and go. Where social network video games diverged from traditional game development was the combination of real-time analytics to continuously optimize game mechanics to drive growth, revenue, and engagement. == Distinct features == The following table outlines common characteristics of social games, mentioned by Björk at the 2010 GCO Games Convention Online: A social network game may employ any of the following features: asynchronous gameplay, which allows rules to be resolved without needing players to play at the same time. gamification, which video game mechanics such as achievements and points are applied to those experienced when playing games in order to motivate and engage users. community, as one of the most distinct features of social video games is in leveraging the player's social network. Quests or game goals may only be possible if a player "shares" with friends connected by the social network hosting the game or gets them to play, as well as "neighbors" or "allies". a lack of victory conditions: there are generally no victory conditions since most developers count on users playing their games often. The game never ends and no one is ever declared winner. Instead, many casual games have "quests" or "missions" for players to complete. This is not true for board game-like social games, such as Scrabble. a virtual currency which players usually must purchase with real-world money. With the in-game currency, players can buy upgrades that would otherwise take much longer to earn through in-game achievements. In many cases, some upgrades are only available with the virtual currency. == Engagement strategies == Since social network games are often less challenging than console games and they have relatively shorter game play, they use different techniques to stretch game play and tools to retain users. Continuous goals: The games assign specific goals for users to achieve. As they advance in the game, the goals become more challenging and time-consuming. They also provide frequent feedback with their performance. Every action will translate towards a certain goal that will be used to attain higher gaming capitals. Gaming capitals: Players are encouraged to earn different badges, trophies, and accolades that indicate their progress and accomplishments. Some achievements are unlocked just by advancing in the game while others may significantly alter the rationale behind the game and require extensive investment from players. The ways of gaining gaming capital are not limited to playing games but the games-related productive activities that are appreciated in the player's social circle too. By accumulating gaming capitals, they provide an intrinsic benefit to gamers as there is an avenue to boost their accomplishment and showcase their expertise of the game. The achievements are visible to their network of friends. Gaming capitals are a way for developers to increase replay value provides extended play time, and players get more value from the game. Motivation for collecting gaming capitals: 1. Legitimization: refers to society's willingness to approve or condone certain behavior. Collecting is about channeling one's materialistic desires into more meaningful pursuits. Game achievements serve a similar purpose, allowing players to justify the hours spent playing the game. 2. Self-extension: Gathering and controlling meaningful objects or experiences can work to gain one an improved sense of self. The collector's goal to complete a collection is symbolically about completing the self too. Events timed to real world: Popular games such as Dragon City and Wild Ones require users to wait a certain time period before their "energy bars" replenish. Without energy, they are unable to conduct any form of action. Gamers are forced to wait and return after their energy replenishes to continue playing. == Monetization == Social network games frequently monetize based on virtual good transactions, but other games are emerging that utilize newer economic models. === Virtual goods === Gamers will be able to purchase in game items like power-ups, avatar accessories, or decorative items users purchase within the game itself. This is realized by monetize products that do not technically exist. Virtual goods account for over 90% of all revenue generated by the world's top social game developers. Designers optimize user experience through additional gameplay, missions, and quests, without having to worry about overhead or unused stock. == Advertising == The following are common ways of advertising in social network games: === Banner advertisements === As banner ads within social networks tend to be where ad response is low, they tend to be priced at bottom-of-the-barrel CPMs of around $2. However, because social games generate so many page views, they are the biggest part of advertising revenue for the social gaming industry. === Video ads === Videos are the ad format with the most revenue per view. They tend to be higher-priced, either by CPMs ($35+ CPM in social games) or cost-per-completed-view. According to studies, video ads result in highest brand recall thus a good return on investment for advertisers. Video ads are shown either in in-game interstitials (e.g. when the game is loading a new screen) or through incentive-based advertising, i.e. you will get either an in-game reward or Facebook credits for watching an advertisement. === Product placement === A brand or product will be injected in a game in some way. Due to the variety of ways in which product placement can be accomplished in any media, and because the category is nascent, this category is not standardized at all, but some examples include branded in-game goods or even in-game quests. For example, in a game where you run a restaurant, you might be asked to collect ingredients to make a Starbucks Frappuccino, and receive in-game rewards for doing so. As these product placement deals are non-standard, they are largely charged with a production fee, which can be $350,000 to $750,000 depending on the type of placement and the popularity of the game. === Lead generation offers === Another form of advertising that is prevalent in many social games are lead generation offers. In this form of advertising, companies, usually from different industries, aim to convince players to sign up for their goods or services and in exchange, players will receive virtual gifts or advance in the game as a reward. === Sponsorship === ==== White label games ==== Applications that are built once, then individualized and licensed again and again. Developer can create a quality app focused on fun while leaving the edge
Read more →