In search of the best AI resume builder? An AI resume builder is software that uses machine learning to help you get more done — it turns a rough idea into a polished result in seconds. When choosing one, weigh output quality, pricing, export formats, and how well it fits the tools you already use. Whether you are a beginner or a pro, the right AI resume builder slots into your workflow and pays for itself fast. We tested the leading options and ranked them by quality, value, and ease of use.
Cloud management
Cloud management refers to the administration and oversight of cloud computing products and services. Public clouds are managed by cloud service providers, which operate the underlying infrastructure such as servers, storage, networking, and data center facilities. Users may also opt to manage their public cloud services with a third-party cloud management tool. Users of public cloud services can generally select from three basic cloud provisioning categories: User self-provisioning: Customers purchase cloud services directly from the provider, typically through a web form or console interface. The customer pays on a per-transaction basis. Advanced provisioning: Customers contract in advance a predetermined amount of resources, which are prepared in advance of service. The customer pays a flat fee or a monthly fee. Dynamic provisioning: The provider allocates resources when the customer needs them, then decommissions them when they are no longer needed. The customer is charged on a pay-per-use basis. Managing a private cloud requires software tools to help create a virtualized pool of compute resources, provide a self-service portal for end users and handle security, resource allocation, tracking and billing. Management tools for private clouds tend to be service driven, as opposed to resource driven, because cloud environments are typically highly virtualized and organized in terms of portable workloads. In hybrid cloud environments, compute, network and storage resources must be managed across multiple domains, so a good management strategy should start by defining what needs to be managed, and where and how to do it. Policies to help govern these domains should include configuration and installation of images, access control, and budgeting and reporting. Access control often includes the use of Single sign-on (SSO), in which a user logs in once and gains access to all systems without being prompted to log in again at each of them. == Characteristics of Cloud Management == Cloud management combines software and technologies in a design for managing cloud environments. Software developers have responded to the management challenges of cloud computing with a variety of cloud management platforms and tools. These tools include native tools offered by public cloud providers as well as third-party tools designed to provide consistent functionality across multiple cloud providers. Administrators must balance the competing requirements of efficient consistency across different cloud platforms with access to different native functionality within individual cloud platforms. The growing acceptance of public cloud and increased multicloud usage is driving the need for consistent cross-platform management. Rapid adoption of cloud services is introducing a new set of management challenges for those technical professionals responsible for managing IT systems and services. Cloud-management platforms and tools should have the ability to provide minimum functionality in the following categories. Functionality can be both natively provided or orchestrated via third-party integration. Provisioning and orchestration: create, modify, and delete resources as well as orchestrate workflows and management of workloads Automation: Enable cloud consumption and deployment of app services via infrastructure-as-code and other DevOps concepts Security and compliance: manage role-based access of cloud services and enforce security configurations Service request: collect and fulfill requests from users to access and deploy cloud resources. Monitoring and logging: collect performance and availability metrics as well as automate incident management and log aggregation Inventory and classification: discover and maintain pre-existing brownfield cloud resources plus monitor and manage changes Cost management and optimization: track and rightsize cloud spend and align capacity and performance to actual demand Migration, backup, and DR: enable data protection, disaster recovery, and data mobility via snapshots and/or data replication Organizations may group these criteria into key use cases including Cloud Brokerage, DevOps Automation, Governance, and Day-2 Life Cycle Operations. Enterprises with large-scale cloud implementations may require more robust cloud management tools which include specific characteristics, such as the ability to manage multiple platforms from a single point of reference, or intelligent analytics to automate processes like application lifecycle management. High-end cloud management tools should also have the ability to handle system failures automatically with capabilities such as self-monitoring, an explicit notification mechanism, and include failover and self-healing capabilities. == Multi-Cloud and Hybrid Cloud Management Challenges == Legacy management infrastructures, which are based on the concept of dedicated system relationships and architecture constructs, are not well suited to cloud environments where instances are continually launched and decommissioned. Instead, the dynamic nature of cloud computing requires monitoring and management tools that are adaptable, extensible and customizable. Cloud computing presents a number of management challenges. Companies using public clouds do not have ownership of the equipment hosting the cloud environment, and because the environment is not contained within their own networks, public cloud customers do not have full visibility or control. Users of public cloud services must also integrate with an architecture defined by the cloud provider, using its specific parameters for working with cloud components. Integration includes tying into the cloud APIs for configuring IP addresses, subnets, firewalls and data service functions for storage. Because control of these functions is based on the cloud provider’s infrastructure and services, public cloud users must integrate with the cloud infrastructure management. Capacity management is a challenge for both public and private cloud environments because end users have the ability to deploy applications using self-service portals. Applications of all sizes may appear in the environment, consume an unpredictable amount of resources, then disappear at any time. A possible solution is profiling the applications impact on computational resources. As result, the performance models allow the prediction of how resource utilization changes according to application patterns. Thus, resources can be dynamically scaled to meet the expected demand. This is critical to cloud providers that need to provision resources quickly to meet a growing demand by their applications. Charge-back—or, pricing resource use on a granular basis—is a challenge for both public and private cloud environments. Charge-back is a challenge for public cloud service providers because they must price their services competitively while still creating profit. Users of public cloud services may find charge-back challenging because it is difficult for IT groups to assess actual resource costs on a granular basis due to overlapping resources within an organization that may be paid for by an individual business unit, such as electrical power. For private cloud operators, charge-back is fairly straightforward, but the challenge lies in guessing how to allocate resources as closely as possible to actual resource usage to achieve the greatest operational efficiency. Exceeding budgets can be a risk. Hybrid cloud environments, which combine public and private cloud services, sometimes with traditional infrastructure elements, present their own set of management challenges. These include security concerns if sensitive data lands on public cloud servers, budget concerns around overuse of storage or bandwidth and proliferation of mismanaged images. Managing the information flow in a hybrid cloud environment is also a significant challenge. On-premises clouds must share information with applications hosted off-premises by public cloud providers, and this information may change constantly. Hybrid cloud environments also typically include a complex mix of policies, permissions and limits that must be managed consistently across both public and private clouds. == Cloud Management Platforms (CMP) == CMPs provide a means for a cloud service customer to manage the deployment and operation of applications and associated datasets across multiple cloud service infrastructures, including both on-premises cloud infrastructure and public cloud service provider infrastructure. In other words, CMPs provide management capabilities for hybrid cloud and multi-cloud environments. A cloud management platform (CMP) provides broad cloud management functionality atop both public cloud provider platforms and private cloud platforms. CMPs manage cloud services and resources that are distributed across multiple cloud platforms. The value of CMPs stands in delivering the maximum level of consistency between platforms without comp
Land of Memories
Land of Memories (Chinese: 机忆之地) is a Chinese science-fiction novel by Shen Yang (沈阳), a professor at Tsinghua University's School of Journalism and Communication. The story revolves around a former neuroscientist trying to recover her memories from the metaverse after suffering amnesia due to an accident. It contains almost 6,000 Chinese characters and was shortened from an AI-generated draft that was 43,000 characters long. The process involved 66 prompts spanning almost three hours. The novel was among 18 submissions that won the level-two prize at the Fifth Jiangsu Youth Science Education and Science Fiction Competition (第五届江苏省青年科普科幻作品大赛). The contest was restricted to participants between the age of 14 and 45 but did not forbid entries generated by AI. One of its organizers reached out to Shen after finding out that the professor had been experimenting with writing science fiction using AI. The judges were not told about the novel's origin in advance. Three of them, out of the six, approved the work. One judge, who had worked with AI models before, recognized that the novel was written by AI and criticized the work for lacking emotional appeal. The organizer who had contacted Shen said the novel's introduction was not bad but the story did not develop well. It would not meet the usual standards for publication. However, he still plans to allow AI-generated submissions in 2024. Fu Ruchu, editorial department director of the People's Literature Publishing House, said the novel was not easily identifiable as AI-generated and applauded its logical consistency. She warned that artificial intelligence could endanger the jobs of fiction writers and cause permanent damage to literary language.
Neural architecture search
Neural architecture search (NAS) is a technique for automating the design of artificial neural networks (ANN), a widely used model in the field of machine learning. NAS has been used to design networks that are on par with or outperform hand-designed architectures. Methods for NAS can be categorized according to the search space, search strategy and performance estimation strategy used: The search space defines the type(s) of ANN that can be designed and optimized. The search strategy defines the approach used to explore the search space. The performance estimation strategy evaluates the performance of a possible ANN from its design (without constructing and training it). NAS is closely related to hyperparameter optimization and meta-learning and is a subfield of automated machine learning (AutoML). == Reinforcement learning == Reinforcement learning (RL) can underpin a NAS search strategy. Barret Zoph and Quoc Viet Le applied NAS with RL targeting the CIFAR-10 dataset and achieved a network architecture that rivals the best manually-designed architecture for accuracy, with an error rate of 3.65, 0.09 percent better and 1.05x faster than a related hand-designed model. On the Penn Treebank dataset, that model composed a recurrent cell that outperforms LSTM, reaching a test set perplexity of 62.4, or 3.6 perplexity better than the prior leading system. On the PTB character language modeling task it achieved bits per character of 1.214. Learning a model architecture directly on a large dataset can be a lengthy process. NASNet addressed this issue by transferring a building block designed for a small dataset to a larger dataset. The design was constrained to use two types of convolutional cells to return feature maps that serve two main functions when convoluting an input feature map: normal cells that return maps of the same extent (height and width) and reduction cells in which the returned feature map height and width is reduced by a factor of two. For the reduction cell, the initial operation applied to the cell's inputs uses a stride of two (to reduce the height and width). The learned aspect of the design included elements such as which lower layer(s) each higher layer took as input, the transformations applied at that layer and to merge multiple outputs at each layer. In the studied example, the best convolutional layer (or "cell") was designed for the CIFAR-10 dataset and then applied to the ImageNet dataset by stacking copies of this cell, each with its own parameters. The approach yielded accuracy of 82.7% top-1 and 96.2% top-5. This exceeded the best human-invented architectures at a cost of 9 billion fewer FLOPS—a reduction of 28%. The system continued to exceed the manually-designed alternative at varying computation levels. The image features learned from image classification can be transferred to other computer vision problems. E.g., for object detection, the learned cells integrated with the Faster-RCNN framework improved performance by 4.0% on the COCO dataset. In the so-called Efficient Neural Architecture Search (ENAS), a controller discovers architectures by learning to search for an optimal subgraph within a large graph. The controller is trained with policy gradient to select a subgraph that maximizes the validation set's expected reward. The model corresponding to the subgraph is trained to minimize a canonical cross entropy loss. Multiple child models share parameters, ENAS requires fewer GPU-hours than other approaches and 1000-fold less than "standard" NAS. On CIFAR-10, the ENAS design achieved a test error of 2.89%, comparable to NASNet. On Penn Treebank, the ENAS design reached test perplexity of 55.8. == Evolution == An alternative approach to NAS is based on evolutionary algorithms, which has been employed by several groups. An Evolutionary Algorithm for Neural Architecture Search generally performs the following procedure. First a pool consisting of different candidate architectures along with their validation scores (fitness) is initialised. At each step the architectures in the candidate pool are mutated (e.g.: 3x3 convolution instead of a 5x5 convolution). Next the new architectures are trained from scratch for a few epochs and their validation scores are obtained. This is followed by replacing the lowest scoring architectures in the candidate pool with the better, newer architectures. This procedure is repeated multiple times and thus the candidate pool is refined over time. Mutations in the context of evolving ANNs are operations such as adding or removing a layer, which include changing the type of a layer (e.g., from convolution to pooling), changing the hyperparameters of a layer, or changing the training hyperparameters. On CIFAR-10 and ImageNet, evolution and RL performed comparably, while both slightly outperformed random search. == Bayesian optimization == Bayesian Optimization (BO), which has proven to be an efficient method for hyperparameter optimization, can also be applied to NAS. In this context, the objective function maps an architecture to its validation error after being trained for a number of epochs. At each iteration, BO uses a surrogate to model this objective function based on previously obtained architectures and their validation errors. One then chooses the next architecture to evaluate by maximizing an acquisition function, such as expected improvement, which provides a balance between exploration and exploitation. Acquisition function maximization and objective function evaluation are often computationally expensive for NAS, and make the application of BO challenging in this context. Recently, BANANAS has achieved promising results in this direction by introducing a high-performing instantiation of BO coupled to a neural predictor. == Hill-climbing == Another group used a hill climbing procedure that applies network morphisms, followed by short cosine-annealing optimization runs. The approach yielded competitive results, requiring resources on the same order of magnitude as training a single network. E.g., on CIFAR-10, the method designed and trained a network with an error rate below 5% in 12 hours on a single GPU. == Multi-objective search == While most approaches solely focus on finding architecture with maximal predictive performance, for most practical applications other objectives are relevant, such as memory consumption, model size or inference time (i.e., the time required to obtain a prediction). Because of that, researchers created a multi-objective search. LEMONADE is an evolutionary algorithm that adopted Lamarckism to efficiently optimize multiple objectives. In every generation, child networks are generated to improve the Pareto frontier with respect to the current population of ANNs. Neural Architect is claimed to be a resource-aware multi-objective RL-based NAS with network embedding and performance prediction. Network embedding encodes an existing network to a trainable embedding vector. Based on the embedding, a controller network generates transformations of the target network. A multi-objective reward function considers network accuracy, computational resource and training time. The reward is predicted by multiple performance simulation networks that are pre-trained or co-trained with the controller network. The controller network is trained via policy gradient. Following a modification, the resulting candidate network is evaluated by both an accuracy network and a training time network. The results are combined by a reward engine that passes its output back to the controller network. == One-shot models == RL or evolution-based NAS require thousands of GPU-days of searching/training to achieve state-of-the-art computer vision results as described in the NASNet, mNASNet and MobileNetV3 papers. To reduce computational cost, many recent NAS methods rely on the weight-sharing idea. In this approach, a single overparameterized supernetwork (also known as the one-shot model) is defined. A supernetwork is a very large Directed Acyclic Graph (DAG) whose subgraphs are different candidate neural networks. Thus, in a supernetwork, the weights are shared among a large number of different sub-architectures that have edges in common, each of which is considered as a path within the supernet. The essential idea is to train one supernetwork that spans many options for the final design rather than generating and training thousands of networks independently. In addition to the learned parameters, a set of architecture parameters are learnt to depict preference for one module over another. Such methods reduce the required computational resources to only a few GPU days. More recent works further combine this weight-sharing paradigm, with a continuous relaxation of the search space, which enables the use of gradient-based optimization methods. These approaches are generally referred to as differentiable NAS and have proven very efficient in exploring the search space of ne
Maschinen Krieger ZbV 3000
Maschinen Krieger (Ma.K ZBV3000), often abbreviated as Ma.K., is a science fiction intellectual property created by Japanese artist and sculptor Kow Yokoyama in the 1980s. It consists of an illustrated series, a line of merchandise comprising display and action figures of mecha characters and a 1985 short film. == History == The franchise originally began as the science fiction series SF3D which ran as monthly installments in the Japanese hobby magazine Hobby Japan from 1982 to 1985. To develop the storyline, Kow Yokoyama collaborated with Hiroshi Ichimura as story editor and Kunitaka Imai as graphic designer. The three creators drew visual inspiration from their combined interest in World War I and World War II armor and aircraft, the American space program and films such as Star Wars, Blade Runner and The Road Warrior. Inspired by the ILM model builders who worked on Star Wars, Yokoyama built the original models from numerous kits including armor, aircraft, and automobiles. He mostly concentrated on powered armor suits, but later included bipedal walking tanks and aircraft with anti-gravity systems. In 1986, there was a dispute with Hobby Japan over the copyright of the series. The magazine dropped SF3D from its line-up of articles and Nitto ceased production of various kits of the series. The matter was tied up in the courts for years until Yokoyama was awarded the full copyright to the series in the 1990s. Yokoyama and Hobby Japan eventually reconciled and restarted their working relationship, ditching the old SF3D name in favor of Maschinen Krieger ZbV3000, otherwise known as Ma.K. == Story == A nuclear World War IV in 2807 kills most of Earth's population and renders the planet uninhabitable. Fifty-two years after the war, a research team from an interstellar union called the Galactic Federation is sent to Earth and discovers that the planet's natural environment has restored itself. The Federation decides to repopulate the planet and sends over colonists to the surface. Cities and towns are eventually reformed over the next 20 years, but this growth attracts the attention of criminals, military deserters, and other lawless elements who wanted to hide on Earth—away from the authorities. A few militias protect the colonists, but the new interlopers often defeat them. Fearing civil unrest and the colonists forming their own government, the Federation gives the Strahl Democratic Republic (SDR) the right to govern the planet in the late 2870s. The SDR sends three police battalions and three Foreign Legion corps to Earth and uses heavy-handed tactics such as travel restrictions and hard labor camps to restore order, which creates resentment amongst the colonists. In response, the colonists create the Earth Independent Provisional Government and declare independence from the SDR. The SDR immediately establishes a puppet government and attempts to quell the uprising. The wealthy colonists hire mercenaries who are descendants of WWIV veterans to form the Independent Mercenary Army (IMA), which is bolstered by the presence of SDR Foreign Legion defectors. They attack the SDR forces and the battle to control Earth begins in 2882. Over the next four years, the SDR and IMA fight each other at several locations worldwide while developing new technology along the way. The war turns up a notch in June 2883 when the IMA deploys a new weapon - the Armored Fighting Suit powered armor - to devastating effect. The SDR eventually builds their own AFS units. In the last SF3D installment published in the December 1986 issue of Hobby Japan, the IMA successfully defeats the new SDR Königs Kröte unmanned command-and-control mecha using a computer virus that also creates a new artificial intelligence system on the moon. == Merchandise == === Model kits === Fan interest from the installments in Hobby Japan resulted in a small Japanese model company, Nitto, securing the license and quickly released 21 injection molded kits from the series during its entire run in the magazine. Most of the Nitto model kits are in 1:20 scale, while others were made in 1:76 and 1:6 scale. Production of the kits stopped with the end of the Hobby Japan features in 1986, but Nitto reissued many of the original kits under the Maschinen Krieger name, albeit with new decals and box art. Some of the original Nitto kits such as the Krachenvogel are highly sought after by collectors. The Nitto models were also the basis for similar offerings from Japanese model companies Wave and ModelKasten. Wave, in particular, is currently producing original-tooled kits of various subjects in the franchise, such as the Armored Fighting Suits powered armor. Smaller companies such as Brick Works and Love Love Garden have made limited resin pilot figures to go with these model kits. At the 2008 Nuremberg Toy Fair in Germany, the Hasegawa company - known mostly for its line of military and civilian vehicles — announced plans to carry the Ma.K license, having successfully branched into pop culture franchises such as Macross. Hasegawa's venture into the franchise came with the release of the Pkf 85 Falke attack craft in March 2009. The company's Ma.K line has since expanded to at least ten kits either 1:35 or 1:20 scale, including a 1:35 Scale Nutrocker tank and the Mk44 humanoid mecha suit from Robot Battle V, a sidestory to the franchise. Wave corporation also has a line of 1/20 models. While Hasegawa largely maintained the yellow-box aesthetic from the older nitto kits, Wave has a more colorful box design. Certain garage kit manufacturers such as Rainbow-Egg are allowed to produce their own line of resin kits and accessories, upon securing special authorization from Yokoyama himself. === Toys === The franchise also contains a line of action and display figures. The Japanese hobby shop and toy company Yellow Submarine and garage kit maker Max Factory released several pre-finished figures in 1:35 and 1:16 scale. MediCom Toys included Chibi Ma.K. figures in their Kubrick line, plus two 1:6 SAFS figures with working lights and fully poseable pilot figures. === Books === Numerous sourcebooks and modeling guides that further flesh out the information in the series have been released. Hobby Japan published a compilation of the first 15 SF3D installments in 1983 and reprinted them in March 2010. Eventually, the magazine re-released all 43 installments in a slipcase compilation called "SF3D Chronicles" in August 2010, which organized the installments into two separate books: "Heaven" featuring articles on aerial models, and "Earth" for ground-based models. Model Graphix followed suit with their own line of sourcebooks, which provide tutorials from Yokoyama on how he makes his figures. Some sourcebooks also have custom decal sets. === Miniature wargaming === In 2019, Slave 2 Gaming gained the license to produce and sell 1:100 scale (15mm) metal and resin war gaming miniatures. This new range of Maschinen Krieger figures was given the name Ma.K in 15mm, so as to not complicate sales with customers, and rebrand the Ma.k name for the miniature wargaming world. The figures are designed and cast in Australia. They are sold exclusively through Slave 2 Gaming at this time due to the license agreement with Sensei Yokoyama. With the production of the miniatures, a set of gaming rules in the works, with the plan is to release all the current Maschinen Krieger models. == Short film == Yokoyama collaborated with Tsuburaya Productions to create a live-action SF3D film using miniatures in 1985. Directed by Shinichi Ohoka from a script penned by co-producer Hisao Ichikura, the 25-minute SF3D Original Video opens with wreckage left from a battle in the Wiltshire wastelands on Christmas Day 2884 before focusing on a badly damaged IMA SAFS unit. The pilot, Cpl Robert Bush (Tristan Hickey), who is still alive, seeks to get his armored suit back and running and leave the battle area, which is under heavy jamming. Seeing two of the SDR's new Nutrocker (Nutcracker) robot hovertanks arrive nearby, Bush tries to hide, but bodily functions give him away. One Nutcracker gives chase and the SAFS AI points out to Bush how to defeat it. He eventually clambers on to the tank, which passes through the rubble of a town and randomly shoots at high places to bring down objects that could snag him. With the SAFS' right arm sheared off by the Nutcracker's laser blasts and snow settling in, Bush is knocked unconscious all night long from the fall while the tank breaks down under the cold. The next day, the SAFS AI wakes up Bush because the Nutcracker is active again and is preparing to kill him. Bush gets up and faces the tank as it charges towards him. However, the Nutcracker gets too close to a cliff that buckles under its weight and Bush fires his laser into the tank's underbelly. The tank plunges into a ravine and explodes. Bush walks away and reestablishes radio contact with his base. It is revealed that the battle was a field test of th
Action model learning
Action model learning (sometimes abbreviated action learning) is an area of machine learning concerned with the creation and modification of a software agent's knowledge about the effects and preconditions of the actions that can be executed within its environment. This knowledge is usually represented in a logic-based action description language and used as input for automated planners. Learning action models is important when goals change. When an agent acted for a while, it can use its accumulated knowledge about actions in the domain to make better decisions. Thus, learning action models differs from reinforcement learning. It enables reasoning about actions instead of expensive trials in the world. Action model learning is a form of inductive reasoning, where new knowledge is generated based on the agent's observations. The usual motivation for action model learning is the fact that manual specification of action models for planners is often a difficult, time-consuming, and error-prone task (especially in complex environments). == Action models == Given a training set E {\displaystyle E} consisting of examples e = ( s , a , s ′ ) {\displaystyle e=(s,a,s')} , where s , s ′ {\displaystyle s,s'} are observations of a world state from two consecutive time steps t , t ′ {\displaystyle t,t'} and a {\displaystyle a} is an action instance observed in time step t {\displaystyle t} , the goal of action model learning in general is to construct an action model ⟨ D , P ⟩ {\displaystyle \langle D,P\rangle } , where D {\displaystyle D} is a description of domain dynamics in action description formalism like STRIPS, ADL or PDDL and P {\displaystyle P} is a probability function defined over the elements of D {\displaystyle D} . However, many state of the art action learning methods assume determinism and do not induce P {\displaystyle P} . In addition to determinism, individual methods differ in how they deal with other attributes of domain (e.g. partial observability or sensoric noise). == Action learning methods == === State of the art === Recent action learning methods take various approaches and employ a wide variety of tools from different areas of artificial intelligence and computational logic. As an example of a method based on propositional logic, we can mention SLAF (Simultaneous Learning and Filtering) algorithm, which uses agent's observations to construct a long propositional formula over time and subsequently interprets it using a satisfiability (SAT) solver. Another technique, in which learning is converted into a satisfiability problem (weighted MAX-SAT in this case) and SAT solvers are used, is implemented in ARMS (Action-Relation Modeling System). Two mutually similar, fully declarative approaches to action learning were based on logic programming paradigm Answer Set Programming (ASP) and its extension, Reactive ASP. In another example, bottom-up inductive logic programming approach was employed. Several different solutions are not directly logic-based. For example, the action model learning using a perceptron algorithm or the multi level greedy search over the space of possible action models. In the older paper from 1992, the action model learning was studied as an extension of reinforcement learning. Nonetheless, further algorithms can be found that operate under different assumptions: FAMA can work even when some observations are missing, and it produces a general (lifted) planning model. It treats learning an action model like a planning problem, making sure the learned model matches the observations given. NOLAM can learn general action models even from noisy or imperfect data. LOCM focuses only on the order of actions in the data, ignoring any details about the states between those actions. The family of safe action model (SAM) learning methods create models that guarantee any plans made with them will actually work in the real world. There's also an extension called N-SAM that can learn action models with numeric conditions and effects. Additionally, numeric action models like N-SAM can be used to improve reinforcement learning (RL) performance through the RAMP algorithm. === Literature === Most action learning research papers are published in journals and conferences focused on artificial intelligence in general (e.g. Journal of Artificial Intelligence Research (JAIR), Artificial Intelligence, Applied Artificial Intelligence (AAI) or AAAI conferences). Despite mutual relevance of the topics, action model learning is usually not addressed in planning conferences like the International Conference on Automated Planning and Scheduling (ICAPS).
Fuzzy cognitive map
A fuzzy cognitive map (FCM) is a cognitive map within which the relations between the elements (e.g. concepts, events, project resources) of a "mental landscape" can be used to compute the "strength of impact" of these elements. Fuzzy cognitive maps were introduced by Bart Kosko. Robert Axelrod introduced cognitive maps as a formal way of representing social scientific knowledge and modeling decision making in social and political systems, then brought in the computation. == Details == Fuzzy cognitive maps are signed fuzzy directed graphs. Spreadsheets or tables are used to map FCMs into matrices for further computation. FCM is a technique used for causal knowledge acquisition and representation, it supports causal knowledge reasoning process and belong to the neuro-fuzzy system that aim at solving decision making problems, modeling and simulate complex systems. Learning algorithms have been proposed for training and updating FCMs weights mostly based on ideas coming from the field of Artificial Neural Networks. Adaptation and learning methodologies used to adapt the FCM model and adjust its weights. Kosko and Dickerson (Dickerson & Kosko, 1994) suggested the Differential Hebbian Learning (DHL) to train FCM. There have been proposed algorithms based on the initial Hebbian algorithm; others algorithms come from the field of genetic algorithms, swarm intelligence and evolutionary computation. Learning algorithms are used to overcome the shortcomings that the traditional FCM present i.e. decreasing the human intervention by suggested automated FCM candidates; or by activating only the most relevant concepts every execution time; or by making models more transparent and dynamic. Fuzzy cognitive maps (FCMs) have gained considerable research interest due to their ability in representing structured knowledge and model complex systems in various fields. This growing interest led to the need for enhancement and making more reliable models that can better represent real situations. A first simple application of FCMs is described in a book of William R. Taylor, where the war in Afghanistan and Iraq is analyzed. In Bart Kosko's book Fuzzy Thinking, several Hasse diagrams illustrate the use of FCMs. As an example, one FCM quoted from Rod Taber describes 11 factors of the American cocaine market and the relations between these factors. For computations, Taylor uses pentavalent logic (scalar values out of {-1,-0.5,0,+0.5,+1}). That particular map of Taber uses trivalent logic (scalar values out of {-1,0,+1}). Taber et al. also illustrate the dynamics of map fusion and give a theorem on the convergence of combination in a related article. While applications in social sciences introduced FCMs to the public, they are used in a much wider range of applications, which all have to deal with creating and using models of uncertainty and complex processes and systems. Examples: In business FCMs can be used for product planning and decision support. In economics, FCMs support the use of game theory in more complex settings. In education for modeling Critical Success Factors of Learning Management Systems. In medical applications to model systems, provide diagnosis, develop decision support systems and medical assessment. In engineering for modeling and control mainly of complex systems and reliability engineering In project planning FCMs help to analyze the mutual dependencies between project resources. In robotics FCMs support machines to develop fuzzy models of their environments and to use these models to make crisp decisions. In computer assisted learning FCMs enable computers to check whether students understand their lessons. In expert systems a few or many FCMs can be aggregated into one FCM in order to process estimates of knowledgeable persons. In IT project management, a FCM-based methodology helps to success modelling, risk analysis and assessment, IT scenarios FCMappers is an international online community for the analysis and the visualization of fuzzy cognitive maps. FCMappers offer support for starting with FCM and also provide a Microsoft Excel-based tool that is able to check and analyse FCMs. The output is saved as Pajek file and can be visualized within third party software like Pajek, Visone, etc. They also offer to adapt the software to specific research needs. Additional FCM software tools, such as Mental Modeler, have recently been developed as a decision-support tool for use in social science research, collaborative decision-making, and natural resource planning.