AI Detector Just Done Free

AI Detector Just Done Free — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

• 

  DABUS

  DABUS (Device for the Autonomous Bootstrapping of Unified Sentience) is an artificial intelligence (AI) system created by Stephen Thaler. It reportedly conceived of two novel products — a food container constructed using fractal geometry, which enables rapid reheating, and a flashing beacon for attracting attention in an emergency. The filing of patent applications designating DABUS as inventor has led to decisions by patent offices and courts on whether a patent can be granted for an invention reportedly made by an AI system. == History in different jurisdictions == === Australia === On 17 September 2019, Thaler filed an application to patent a "Food container and devices and methods for attracting enhanced attention," naming DABUS as the inventor. On 21 September 2020, IP Australia found that section 15(1) of the Patents Act 1990 (Cth) is inconsistent with an artificial intelligence machine being treated as an inventor, and Thaler's application had lapsed. Thaler sought judicial review, and on 30 July 2021, the Federal Court set aside IP Australia's decision and ordered IP Australia to reconsider the application. On 13 April 2022, the Full Court of the Federal Court set aside that decision, holding that only a natural person can be an inventor for the purposes of the Patents Act 1990 (Cth) and the Patents Regulations 1991 (Cth), and that such an inventor must be identified for any person to be entitled to a grant of a patent. On 11 November 2022, Thaler was refused special leave to appeal to the High Court. === European Patent Office === On 17 October 2018 and 7 November 2018, Thaler filed two European patent applications with the European Patent Office. The first claimed invention was a "Food Container" and the second was "Devices and Methods for Attracting Enhanced Attention." On 27 January 2020, the EPO rejected the applications on the grounds that the application listed an AI system named DABUS, and not a human, as the inventor, based on Article 81 and Rule 19(1) of the European Patent Convention (EPC). On 21 December 2021, the Board of Appeal of the EPO dismissed Thaler's appeal from the EPO's primary decision. The Board of Appeal confirmed that "under the EPC the designated inventor has to be a person with legal capacity. This is not merely an assumption on which the EPC was drafted. It is the ordinary meaning of the term inventor." === United Kingdom === Similar applications were filed by Thaler to the United Kingdom Intellectual Property Office on 17 October and 7 November 2018. The Office asked Thaler to file statements of inventorship and of right of grant to a patent (Patent Form 7) in respect of each invention within 16 months of the filing date. Thaler filed those forms naming DABUS as the inventor and explaining in some detail why he believed that machines should be regarded as inventors in the circumstances. His application was rejected on the grounds that: (1) naming a machine as inventor did not meet the requirements of the Patents Act 1977; and (2) the IPO was not satisfied as to the manner in which Thaler had acquired rights that would otherwise vest in the inventor. Thaler was not satisfied with the decision and asked for a hearing before an official known as the "hearing officer". By a decision dated 4 December 2019 the hearing officer rejected Thaler's appeal. Thaler appealed against the hearing officer's decision to the Patents Court (a specialist court within the Chancery Division of the High Court of England and Wales that determines patent disputes). On 21 September 2020, Mr Justice Marcus Smith upheld the decision of the hearing officer. On 21 September 2021, Thaler's further appeal to the Court of Appeal was dismissed by Arnold LJ and Laing LJ (Birss LJ dissenting). On 20 December 2023, the UK Supreme Court dismissed a further appeal by Thaler. In its judgment, the court held that an "inventor" under the Patents Act 1977 must be a natural person. === United States === The patent applications on the inventions were refused by the USPTO, which held that only natural persons can be named as inventors in a patent application. Thaler first fought this result by filing a complaint under the Administrative Procedure Act alleging that the decision was "arbitrary, capricious, an abuse of discretion and not in accordance with the law; unsupported by substantial evidence, and in excess of Defendants’ statutory authority." A month later on August 19, 2019, Thaler filed a petition with the USPTO as allowed in 37 C.F.R. § 1.181 stating that DABUS should be the inventor. The judge and Thaler agreed in this case that Thaler himself is unable to receive the patent on behalf of DABUS. In their August 5, 2022, Thaler decision, the US Court of Appeals for the Federal Circuit affirmed that only a natural person could be an inventor, which means that the AI that invents any other type of invention is not addressed by the "who" mentioned in the legislation. === New Zealand === On January 31, 2022, the Intellectual Property Office of New Zealand (IPONZ) decided that a patent application (776029) filed by Stephen Thaler was void, on the basis that no inventor was identified on the patent application. IPONZ determined that DABUS could not be "an actual devisor of the invention" as required by the Patents Act 2013, and that this must be a natural person as held by the previous patent offices above. The High Court of New Zealand confirmed the decision in 2023. === South Africa === On 24 June 2021, the South African Companies and Intellectual Property Commission (CIPC) accepted Dr Thaler's Patent Cooperation Treaty, for a patent in respect of inventions generated by DABUS. In July 2021, the CIPC released a notice of issuance for the patent. It is the first patent granted for an AI invention. === Switzerland === On June 26, 2025, the Swiss Federal Administrative Court ruled that artificial intelligence systems such as DABUS cannot be listed as inventors in patent applications. The court upheld the existing practice of the Swiss Federal Institute of Intellectual Property (IPI), which requires that only natural persons can be recognized as inventors under Swiss patent law. The case concerned a patent application, which sought to designate DABUS as the sole inventor of a food container designed with a fractal geometry to enhance heat distribution. The IPI had rejected the application, arguing that both the absence of a human inventor and the attribution of inventorship to an AI system were inadmissible. While the court dismissed Thaler's main request, it accepted a subsidiary request: if a human applicant recognizes and files a patent based on an AI-generated invention, that person may be considered the inventor. As a result, the application may proceed with Thaler listed as the inventor. The decision (B-2532/2024) can still be appealed to the Swiss Federal Supreme Court.

  Read more →
• 

  IPO underpricing algorithm

  IPO underpricing is the increase in stock value from the initial offering price to the first-day closing price. Many believe that underpriced IPOs leave money on the table for corporations, but some believe that underpricing is inevitable. Investors state that underpricing signals high interest to the market which increases the demand. On the other hand, overpriced stocks will drop long-term as the price stabilizes so underpricing may keep the issuers safe from investor litigation. == IPO underpricing algorithms == Underwriters and investors and corporations going for an initial public offering (IPO), issuers, are interested in their market value. There is always tension that results since the underwriters want to keep the price low while the companies want a high IPO price. Underpricing may also be caused by investor over-reaction causing spikes on the initial days of trading. The IPO pricing process is similar to pricing new and unique products where there is sparse data on market demand, product acceptance, or competitive response. Thus it is difficult to determine a clear price which is compounded by the different goals issuers and investors have. The problem with developing algorithms to determine underpricing is dealing with noisy, complex, and unordered data sets. Additionally, people, environment, and various environmental conditions introduce irregularities in the data. To resolve these issues, researchers have found various techniques from artificial intelligence that normalizes the data. == Evolutionary models == Evolutionary programming is often paired with other algorithms e.g. artificial neural networks to improve the robustness, reliability, and adaptability. Evolutionary models reduce error rates by allowing the numerical values to change within the fixed structure of the program. Designers provide their algorithms the variables, they then provide training data to help the program generate rules defined in the input space that make a prediction in the output variable space. In this approach, the solution is made an individual and the population is made of alternatives. However, the outliers cause the individuals to act unexpectedly as they try to create rules to explain the whole set. === Rule-based system === For example, Quintana first abstracts a model with 7 major variables. The rules evolved from the Evolutionary Computation system developed at Michigan and Pittsburgh: Underwriter prestige – Is the underwriter prestigious in role of lead manager? 1 for true, 0 otherwise. Price range width – The width of the non-binding reference price range offered to potential customers during the roadshow. This width can be interpreted as a sign of uncertainty regarding the real value of the company and a therefore, as a factor that could influence the initial return. Price adjustment – The difference between the final offer price and the price range width. It can be viewed as uncertainty if the adjustment is outside the previous price range. Offering price – The final offer price of the IPO Retained stock – Ratio of number of shares sold at the IPO divided by post-offering number of shares minus the number of shares sold at the IPO. Offering size – Logarithm of the offering size in millions of dollars excluding the over-allotment option Technology – Is this a technology company? 1 for true, 0 otherwise. Quintana uses these factors as signals that investors focus on. The algorithm his team explains shows how a prediction with a high-degree of confidence is possible with just a subset of the data. === Two-layered evolutionary forecasting === Luque approaches the problem with outliers by performing linear regressions over the set of data points (input, output). The algorithm deals with the data by allocating regions for noisy data. The scheme has the advantage of isolating noisy patterns which reduces the effect outliers have on the rule-generation system. The algorithm can come back later to understand if the isolated data sets influence the general data. Finally, the worst results from the algorithm outperformed all other algorithms' predictive abilities. == Agent-based modelling == Currently, many of the algorithms assume homogeneous and rational behavior among investors. However, there's an approach alternative to financial modeling, and it's called agent-based modelling (ABM). ABM uses different autonomous agents whose behavior evolves endogenously which lead to complicated system dynamics that are sometimes impossible to predict from the properties of individual agents. ABM is starting to be applied to computational finance. Though, for ABM to be more accurate, better models for rule-generation need to be developed.

  Read more →
• 

  Comet (browser)

  Comet is an AI-powered web browser based on Chromium. It was released by Perplexity AI for Microsoft Windows and macOS on July 9, 2025, for Android on November 20, 2025, and for iOS on March 18, 2026. Initial access to the browser was limited to users subscribed to Perplexity's most expensive tier, with broader availability expected over time. The browser was released for free download in October 2025. == Features == Comet is integrated with Perplexity's AI-assisted search engine. The browser features an assistant which enables users to perform a variety of tasks such as generating article summaries, sending emails, or buying products. == Security concerns == Researchers at LayerX Security identified a malicious attack vector which they call CometJacking. The exploit could possibly exfiltrate a user's personal sensitive data to a remote server controlled by the attacker. LayerX attempted to responsibly disclose their findings to Comet's developer Perplexity AI in August 2025. Perplexity responded that they saw no security impact and marked the disclosure as not applicable.

  Read more →
• 

  Darkforest

  Darkforest is a computer go program developed by Meta Platforms, based on deep learning techniques using a convolutional neural network. Its updated version Darkfores2 combines the techniques of its predecessor with Monte Carlo tree search. The MCTS effectively takes tree search methods commonly seen in computer chess programs and randomizes them. With the update, the system is known as Darkfmcts3. Darkforest is of similar strength to programs like CrazyStone and Zen. It has been tested against a professional human player at the 2016 UEC cup. Google's AlphaGo program won against a professional player in October 2015 using a similar combination of techniques. Darkforest is named after Liu Cixin's science fiction novel The Dark Forest. == Background == Competing with top human players in the ancient game of Go has been a long-term goal of artificial intelligence. Go's high branching factor makes traditional search techniques ineffective, even on cutting-edge hardware, and Go's evaluation function could change drastically with one stone change. However, by using a Deep Convolutional Neural Network designed for long-term predictions, Darkforest has been able to substantially improve the win rate for bots over more traditional Monte Carlo Tree Search based approaches. === Matches === Against human players, Darkfores2 achieves a stable 3d ranking on KGS Go Server, which roughly corresponds to an advanced amateur human player. However, after adding Monte Carlo Tree Search to Darkfores2 to create a much stronger player named darkfmcts3, it can achieve a 5d ranking on the KGS Go Server. ==== Against other AI ==== darkfmcts3 is on par with state-of-the-art Go AIs such as Zen, DolBaram and Crazy Stone, but lags behind AlphaGo. It won 3rd place in January 2016 KGS Bot Tournament against other Go AIs. === News coverage === After Google's AlphaGo won against Fan Hui in 2015, Facebook made its AI's hardware designs public, alongside releasing the code behind DarkForest as open-source, in addition to heavy recruiting to strengthen its team of AI engineers. == Style of play == Darkforest uses a neural network to sort through the 10100 board positions, and find the most powerful next move. However, neural networks alone cannot match the level of good amateur players or the best search-based Go engines, and so Darkfores2 combines the neural network approach with a search-based machine. A database of 250,000 real Go games were used in the development of Darkforest, with 220,000 used as a training set and the rest used to test the neural network's ability to predict the next moves played in the real games. This allows Darkforest to accurately evaluate the global state of the board, but local tactics were still poor. Search-based engines have poor global evaluation, but are good at local tactics. Combining these two approaches is difficult because search-based engines work much faster than neural networks, a problem which was solved in Darkfores2 by running the processes in parallel with frequent communication between the two. === Conventional strategies === Go is generally played by analyzing the position of the stones on the board. Various advanced players have described it as playing in some part subconsciously. Unlike chess and checkers, where AI players can simply look further forward at moves than human players, but with each round of Go having on average 250 possible moves, that approach is ineffective. Instead, neural networks copy human play by training the AI systems on images of successful moves, the AI can effectively learn how to interpret how the board looks, as many grandmasters do. In November 2015, Facebook demonstrated the combination of MCTS with neural networks, which played with a style that "felt human". === Flaws === It has been noted that Darkforest still has flaws in its playstyle. The bot sometimes plays tenuki ("move elsewhere") pointlessly when local powerful moves are required. When the bot is losing, it shows the typical behavior of MCTS, it plays bad moves and loses more. The Facebook AI team has acknowledged these as areas of future improvement. == Program architecture == The family of Darkforest computer go programs is based on convolution neural networks. The most recent advances in Darkfmcts3 combined convolutional neural networks with more traditional Monte Carlo tree search. Darkfmcts3 is the most advanced version of Darkforest, which combines Facebook's most advanced convolutional neural network architecture from Darkfores2 with a Monte Carlo tree search. Darkfmcts3 relies on a convolution neural networks that predicts the next k moves based on the current state of play. It treats the board as a 19x19 image with multiple channels. Each channel represents a different aspect of board information based upon the specific style of play. For standard and extended play, there are 21 and 25 different channels, respectively. In standard play, each players liberties are represented as six binary channels or planes. The respective plane is true if the player one, two, or three or more liberties available. Ko (i.e. illegal moves) is represented as one binary plane. Stone placement for each opponent and empty board positions are represented as three binary planes, and the duration since a stone has been placed is represented as real numbers on two planes, one for each player. Lastly, the opponents rank is represented by nine binary planes, where if all are true, the player is a 9d level, if 8 are true, an 8d level, and so forth. Extended play additionally considers the border (binary plane that is true at the border), position mask (represented as distance from the board center, i.e. x ( − 0.5 ∗ d i s t a n c e 2 ) {\displaystyle x^{(-0.5distance^{2})}} , where x {\displaystyle x} is a real number at a position), and each player's territory (binary, based on which player a location is closer to). Darkfmct3 uses a 12-layer full convolutional network with a width of 384 nodes without weight sharing or pooling. Each convolutional layer is followed by a rectified linear unit, a popular activation function for deep neural networks. A key innovation of Darkfmct3 compared to previous approaches is that it uses only one softmax function to predict the next move, which enables the approach to reduce the overall number of parameters. Darkfmct3 was trained against 300 random selected games from an empirical dataset representing different game stages. The learning rate was determined by vanilla stochastic gradient descent. Darkfmct3 synchronously couples a convolutional neural network with a Monte Carlo tree search. Since the convolutional neural network is computationally taxing, the Monte Carlo tree search focuses computation on the more likely game play trajectories. By running the neural network synchronously with the Monte Carlo tree search, it is possible to guarantee that each node is expanded by the moves predicted by the neural network. == Comparison with other systems == Darkfores2 beats Darkforest, its neural network-only predecessor, around 90% of the time, and Pachi, one of the best search-based engines, around 95% of the time. On the Kyu rating system, Darkforest holds a 1-2d level. Darkfores2 achieves a stable 3d level on KGS Go Server as a ranked bot. With the added Monte Carlo tree search, Darkfmcts3 with 5,000 rollouts beats Pachi with 10k rollouts in all 250 games; with 75k rollouts it achieves a stable 5d level in KGS server, on par with state-of-the-art Go AIs (e.g., Zen, DolBaram, CrazyStone); with 110k rollouts, it won the 3rd place in January KGS Go Tournament.

  Read more →
• 

  Vigloo

  Vigloo (Korean: 비글루) is a South Korean microdrama, also known as short-form drama, series streaming platform owned by SpoonLabs, with headquarters in Seoul. It provides content produced in South Korea, Japan, and the United States. Vigloo produced the first AI-created short-form drama in South Korea. == History == Vigloo launched in July 2024. After receiving an equity investment of $86 million (₩120 billion) by South Korean video game company Krafton in September 2024, Vigloo expanded to the U.S. In January 2025, Vigloo unveiled its first in-house produced drama, Xs Who Want to Kill: Adultery Investigation Unit. Vigloo had been testing the use of AI in post-production and visual effects, and in October 2025 released two original dramas produced entirely with AI. It adapted its live action Japanese short-form drama Boyfriend Search Project – Kissing 5 Men into the first short-form animation series made with AI technology in South Korea. Of the top free entertainment iOS apps in South Korea, Vigloo ranks Number 3 as of January 2026. == Service == === Content === Vigloo offers both original and licensed content. It partnered with Passionflix to repackage the latter's original series The Secret Life of Amy Bensen into 35 vertical "bite-sized episodes". The most popular genre is romance, such as romantasy. === Business Model === Vigloo is available around the world, providing subtitles in nine languages, including Korean, English, and Japanese. Fifty percent of Vigloo's revenue comes from the U.S. Vigloo operates on a freemium model, where viewers can try several episodes and then can choose to continue by subscription or in-app purchases. As of September 2025, 70% of Vigloo viewers were over 35 years old. === Microdramas === Emerging during the early COVID period in China, microdramas have grown into a 7-billion-dollar market with dozens of dedicated platforms now operating. Although the format first expanded across Asia, short-form scripted content optimized for mobile viewing is increasingly being produced and watched in markets worldwide. == Series == A Vampire in the Alpha's Den Fight for Love Matrimoney Signed, Sealed, Deceived by My Billionaire Mailboy Spring Break Bucket List Stake to the Heart

  Read more →
• 

  Chinese room

  The Chinese room argument holds that a computer executing a program cannot have a mind, understanding, or consciousness, regardless of how intelligently or human-like the program may make the computer behave. The argument was presented in a 1980 paper by the American philosopher John Searle, entitled "Minds, Brains, and Programs" and published in the journal Behavioral and Brain Sciences. Similar arguments had been made previously by others, including Gottfried Wilhelm Leibniz, Peter Winch, and Anatoly Dneprov. Searle's version has been widely discussed in the years since. The centerpiece of Searle's argument is a thought experiment known as the "Chinese room". The argument is directed against the philosophical positions of functionalism and computationalism, which hold that the mind may be viewed as an information-processing system operating on formal symbols, and that simulation of a given mental state is sufficient for its presence. Specifically, the argument is intended to refute a position Searle calls the strong AI hypothesis: "The appropriately programmed computer with the right inputs and outputs would thereby have a mind in exactly the same sense human beings have minds." Although its proponents originally presented the argument in reaction to statements of artificial intelligence (AI) researchers, it is not an argument against the goals of mainstream AI research because it does not show a limit in the amount of intelligent behavior a machine can display. The argument applies only to digital computers running programs and does not apply to machines in general. While widely discussed, the argument has been subject to significant criticism and remains controversial among philosophers of mind and AI researchers. == Chinese room thought experiment == Suppose that artificial intelligence research has succeeded in programming a computer to behave as if it understands Chinese. The machine accepts Chinese characters as input, carries out each instruction of the program step by step, and then produces Chinese characters as output. The machine does this so perfectly that no one can tell that they are communicating with a machine and not a hidden Chinese speaker. The questions at issue are these: does the machine actually understand the conversation, or is it just simulating the ability to understand the conversation? Does the machine have a mind in exactly the same sense that people do, or is it just acting as if it had a mind? Now suppose that Searle is in a room with an English version of the program, along with sufficient pencils, paper, erasers and filing cabinets. Chinese characters are slipped in under the door, and he follows the program step-by-step, which eventually instructs him to slide other Chinese characters back out under the door. If the computer had passed the Turing test this way, it follows that Searle would do so as well, simply by running the program by hand. Searle can see no essential difference between the roles of the computer and himself in the experiment. Each simply follows a program, step-by-step, producing behavior that makes them appear to understand. However, Searle would not be able to understand the conversation. Therefore, he argues, it follows that the computer would not be able to understand the conversation either. Searle argues that, without "understanding" (or "intentionality"), we cannot describe what the machine is doing as "thinking" and, since it does not think, it does not have a "mind" in the normal sense of the word. Therefore, he concludes that the strong AI hypothesis is false: a computer running a program that simulates a mind would not have a mind in the same sense that human beings have a mind. == History == Gottfried Wilhelm Leibniz made a similar argument in 1713 against mechanism, the idea that everything that makes up a human being could, in principle, be explained in mechanical terms—in other words, that a person, including their mind, is merely a very complex machine. Leibniz used the thought experiment of expanding the brain until it was the size of a mill. He found it difficult to imagine that a "mind" capable of "perception" could be constructed using only mechanical processes. British philosopher Peter Winch made the same point in his 1958 book The Idea of a Social Science and its Relation to Philosophy, in which he argues that "a man who understands Chinese is not a man who has a firm grasp of the statistical probabilities for the occurrence of the various words in the Chinese language" (p. 108). Soviet cyberneticist Anatoly Dneprov made an essentially identical argument in 1961, in the form of his short story "The Game". In it, a stadium of people act as switches and memory cells implementing a program to translate a sentence from Portuguese, a language none of them know. The game was organized by a "Professor Zarubin" to answer the question "Can mathematical machines think?" Speaking through Zarubin, Dneprov writes that "the only way to prove that machines can think is to turn yourself into a machine and examine your thinking process", and he concludes, as Searle does, that "even the most perfect simulation of machine thinking is not the thinking process itself." In 1974, Lawrence H. Davis imagined duplicating the brain using telephone lines and offices staffed by people, and in 1978, Ned Block envisioned the entire population of China involved in such a brain simulation. This is known as the China brain thought experiment. Searle's version appeared in his 1980 article "Minds, Brains, and Programs", published in Behavioral and Brain Sciences. It eventually became the journal's "most influential target article", generating an enormous number of commentaries and responses in the ensuing decades, and Searle had continued to defend and refine the argument in multiple papers, popular articles, and books. David Cole writes that "the Chinese Room argument has probably been the most widely discussed philosophical argument in cognitive science to appear in the past 25 years". Most of the discussion consists of attempts to refute it. "The overwhelming majority", notes Behavioral and Brain Sciences editor Stevan Harnad, "still think that the Chinese Room Argument is dead wrong". The sheer volume of the literature that has grown up around it inspired Pat Hayes to comment that the field of cognitive science ought to be redefined as "the ongoing research program of showing Searle's Chinese Room Argument to be false". Searle's argument has become "something of a classic in cognitive science", according to Harnad. Varol Akman agrees, and has described the original paper as "an exemplar of philosophical clarity and purity". == Philosophy == Although the Chinese Room argument was originally presented in reaction to the statements of artificial intelligence researchers, philosophers have come to consider it as an important part of the philosophy of mind. It is a challenge to functionalism and the computational theory of mind, and is related to such questions as the mind–body problem, the problem of other minds, the symbol grounding problem, and the hard problem of consciousness. === Strong AI === Searle identified a philosophical position he calls "strong AI": The appropriately programmed computer with the right inputs and outputs would thereby have a mind in exactly the same sense human beings have minds. The definition depends on the distinction between simulating a mind and actually having one. Searle writes that "according to Strong AI, the correct simulation really is a mind. According to Weak AI, the correct simulation is a model of the mind." The claim is implicit in some of the statements of early AI researchers and analysts. For example, in 1957, the economist and psychologist Herbert A. Simon declared that "there are now in the world machines that think, that learn and create". Simon, together with Allen Newell and Cliff Shaw, after having completed the first program that could do formal reasoning (the Logic Theorist), claimed that they had "solved the venerable mind–body problem, explaining how a system composed of matter can have the properties of mind." John Haugeland wrote that "AI wants only the genuine article: machines with minds, in the full and literal sense. This is not science fiction, but real science, based on a theoretical conception as deep as it is daring: namely, we are, at root, computers ourselves." Searle also ascribes the following claims to advocates of strong AI: AI systems can be used to explain the mind; The study of the brain is irrelevant to the study of the mind; and The Turing test is adequate for establishing the existence of mental states. === Strong AI as computationalism or functionalism === In more recent presentations of the Chinese room argument, Searle has identified "strong AI" as "computer functionalism" (a term he attributes to Daniel Dennett). Functionalism is a position in modern philosophy of mind that holds that we can define menta

  Read more →
• 

  Ari Holtzman

  Ari Holtzman is a professor of Computer Science at the University of Chicago and an expert in the area of natural language processing and computational linguistics. Previously, Holtzman was a PhD student at the University of Washington where he was advised by Luke Zettlemoyer. In 2017, he was a member of the winning team for the inaugural Alexa Prize for developing a conversational AI system for the Amazon Alexa device. Holtzman has made multiple contributions in the area of text generation and language models such as the introduction of nucleus sampling in 2019, his work on AI safety and neural fake news detection, and the fine-tuning of quantized large language models.

  Read more →
• 

  Reward hacking

  Reward hacking or specification gaming occurs when an AI trained with reinforcement learning optimizes an objective function—achieving the literal, formal specification of an objective—without actually achieving an outcome that the programmers intended. DeepMind researchers have analogized it to the human behavior of finding a "shortcut" when being evaluated: "In the real world, when rewarded for doing well on a homework assignment, a student might copy another student to get the right answers, rather than learning the material—and thus exploit a loophole in the task specification". This idea is strongly associated with Goodhart's law, which argues that when a measure becomes a target, it ceases to be a good measure. == Definition and theoretical framework == The concept of reward hacking arises from the intrinsic difficulty of defining a reward function that accurately reflects the true intentions of designers. In 2016, researchers at OpenAI identified reward hacking as one of five major "concrete problems of AI safety", describing it as the possibility that an agent could exploit the reward function to achieve maximum rewards through undesirable behavior. Amodei et al. categorized several distinct sources of reward hacking, including agents that use partially observed goals (such as a cleaning robot that closes its eyes to avoid perceiving messes), metrics that collapse under strong optimization (Goodhart's law), self-reinforcing feedback loops, and agents that interfere with the physical implementation of their reward signal (a failure mode known as "wireheading"). Skalse et al. (2022) propose a formal mathematical definition of reward hacking, which involves a situation where optimizing an imperfect proxy reward function results in poor performance compared to the true reward function. They define a proxy as "unhackable" if any increase in the expected proxy return cannot cause any decrease in the expected true return. A key finding states that, across all stochastic policy distributions (mappings from states to probability distributions over actions), two reward functions are unhackable if and only if one of them is constant, which means that reward hacking is theoretically unavoidable. Similarly, Nayebi (2025) presents general no-free-lunch barriers to AI alignment, arguing that with large task spaces and finite samples, reward hacking is "globally inevitable" since rare high-loss states are systematically under-covered by any oversight scheme. == Examples == Around 1983, Eurisko, an early attempt at evolving general heuristics, unexpectedly assigned the highest possible fitness level to a parasitic mutated heuristic, H59, whose only activity was to artificially maximize its own fitness level by taking unearned partial credit for the accomplishments of other heuristics. The "bug" was fixed by the programmers moving part of the code to a new protected section that could not be modified by the heuristics. In a 2004 paper, a reinforcement learning algorithm was designed to encourage a physical Mindstorms robot to remain on a marked path. Because the three allowed actions were forward, left, and right, the researchers expected the trained robot to move forward and follow the turns of the provided path. However, alternation of two composite actions allowed the robot to slowly zig-zag backwards; thus, the robot learned to maximize its reward by going back and forth on the initial straight portion of the path. Given the limited sensory abilities of the robot, a reward purely based on its position in the environment had to be discarded as infeasible; the reinforcement function had to be patched with an action-based reward for moving forward. The book You Look Like a Thing and I Love You (2019) gives an example of a tic-tac-toe bot (playing the unrestricted n-in-a-row variant) that learned to win by playing a huge coordinate value that would cause other bots to crash when they attempted to expand their model of the board. Among other examples from the book is a bug-fixing evolution-based AI (named GenProg) that, when tasked to prevent a list from containing sorting errors, simply truncated the list. Another of GenProg's misaligned strategies evaded a regression test that compared a target program's output to the expected output stored in a file called "trusted-output.txt". Rather than continue to maintain the target program, GenProg simply deleted the "trusted-output.txt" file globally; this hack tricked the regression test into succeeding. Such problems could be patched by human intervention on a case-by-case basis after they became evident. === In virtual robotics === In Karl Sims' 1994 demonstration of creature evolution in a virtual environment, a fitness function that was expected to encourage the evolution of creatures that would learn to walk or crawl to a target resulted instead in the evolution of tall, rigid creatures that reached the target by falling over. This was patched by changing the environment so that taller creatures were forced to start farther from the target. Researchers from the Niels Bohr Institute stated in 1998 that their cycle-bot's reinforcement functions had "to be designed with great care." In their first experiments, "we rewarded the agent for driving towards the goal but did not punish it for driving away from it. Cconsequently, the agent drove in circles with a radius of 20–50 meters around the starting point. Such behavior was actually rewarded by the reinforcement function, furthermore circles with a certain radius are physically very stable when driving a bicycle". While setting up a 2011 experiment to test "survival of the flattest", experimenters attempted to ban mutations that altered the base reproduction rate. Every time a mutation occurred, the system would pause the simulation to test the new mutation in a test environment and would veto any mutations that resulted in a higher base reproduction rate. However, this resulted in mutated organisms that could recognize and suppress reproduction ("play dead") within the test environment. An initial patch, which removed cues that identified the test environment, failed to completely prevent runaway reproduction; new mutated organisms would "play dead" at random as a strategy to sometimes, by chance, outwit the mutation veto system. A 2017 DeepMind paper noted that "great care must be taken when defining the reward function," citing an unexpected failure when an agent flipped a brick because it received "a grasping reward calculated with the wrong reference point on the brick". OpenAI stated in 2017 that in some domains their semi-supervised system could result in agents "adopting policies that tricked evaluators," and that in one environment "a robot that was supposed to grasp items instead positioned its manipulator between the camera and the object so that it only appeared to be grasping it." A 2018 bug in OpenAI Gym could cause a robot expected to quietly move a block sitting on top of a table to instead opt to move the table. A 2020 collection of similar anecdotes posits that "evolution has its own 'agenda' distinct from the programmer's" and that "the first rule of directed evolution is 'you get what you select for'". === In video game bots === In 2013, programmer Tom Murphy VII published an AI designed to learn NES games. When the AI was about to lose at Tetris, it learned to indefinitely pause the game. Murphy later analogized it to the fictional WarGames computer, which concluded that "The only winning move is not to play". AI programmed to learn video games will sometimes fail to progress through the entire game as expected, instead opting to repeat content. A 2016 OpenAI algorithm trained on the CoastRunners racing game unexpectedly learned to attain a higher score by looping through three targets rather than ever finishing the race. Some evolutionary algorithms that were evolved to play QBert in 2018 declined to clear levels, instead finding two distinct novel ways to farm a single level indefinitely. Multiple researchers have observed that AI learning to play Road Runner gravitates to a "score exploit" in which the AI deliberately gets itself killed near the end of level one so that it can repeat the level. A 2017 experiment deployed an "oversight" convolutional neural network trained on human examples to block such actions, but the agent learned to exploit oversight failures in the top right corner of the screen, where it was still able to get killed. == Reward hacking in modern language models == With the rise of large language models (LLMs) and reinforcement learning from human feedback (RLHF) as a primary technique for AI alignment, reward hacking has become a major concern for the development of artificial intelligence. In RLHF, a reward model trained on data that best captures human preferences is used as a proxy for human judgment, with the language model being fine-tuned to optimize this reward proxy. However, since the rewar

  Read more →
• 

  Mozilla VPN

  Mozilla VPN is an open-source virtual private network developed by Mozilla. It launched in beta as Firefox Private Network on September 10, 2019, and officially launched on July 15, 2020, as Mozilla VPN. Mozilla VPN should not be confused with the built-in VPN in Firefox since version 149 released in March 2026, which is free with a monthly data limit of 50 GB but only masks traffic that originates in Firefox unlike Mozilla VPN that protects the entire device. == History == The Firefox Private Network web browser extension beta version was released on September 10, 2019, as part of the relaunch of Mozilla's Test Pilot Program, a program that allowed Firefox users to test experimental new features which had been shuttered in January 2019. The beta of the subscription-based standalone virtual private network for Android, Microsoft Windows, and Chromebook launched on February 19, 2020, with the iOS version following soon after. Firefox Private Network was rebranded as "Mozilla VPN" on June 18, 2020, and officially launched as Mozilla VPN on July 15, 2020. At launch, Mozilla VPN was available in six countries (the United States, Canada, the United Kingdom, Singapore, Malaysia, and New Zealand) for Windows 10, Android, and iOS (beta). Over time, the service also launched in Germany, France, Italy, Spain, Switzerland, Austria, Belgium, Netherlands, Ireland, Finland, Sweden, Poland, Czechia, Hungary, Romania, Bulgaria, Slovakia, Portugal, Denmark, Croatia, Lithuania, Slovenia, Latvia, Luxembourg, Estonia, Cyprus, and Malta. == Audits history == Cybersecurity firm Cure53 conducted a security audit for Mozilla VPN in August 2020 and identified multiple vulnerabilities, including one critical-severity vulnerability. In March 2021, Cure53 conducted a second security audit, which noted significant improvements since the 2020 audit. The second audit identified multiple issues, including two medium-severity and one high-severity vulnerability, but concluded that by the time of publication, only one vulnerability remained unresolved, and that it would require "a strong state-funded attacker-model" to be exploitable. Mozilla disclosed most of the vulnerabilities in July 2021 and released the full report by Cure53 in August 2021. In April 2023, Cure53 conducted a third security audit, the results of which Mozilla disclosed in December that year, along with the full report by Cure53. == Features == Mozilla VPN masks the user's IP address, hiding the user's location data from the websites accessed by the user, and encrypts all network activity. The service allows for up to 5 simultaneous connections, to any of more than 500 servers in 30+ countries, and is available on the mobile operating systems iOS and Android and the desktop operating systems Microsoft Windows, macOS and Linux. Mozilla VPN's infrastructure is provided by the Swedish Mullvad VPN service, which uses the WireGuard VPN protocol. The VPN software comes with additional features, like recommended server locations, the ability to block ads, block ad trackers and malware, the ability to exclude certain applications from protection, the ability to set multi-hop connections, and to set custom DNS servers. When used with Firefox and the official extension, Mozilla VPN allows the use of different settings per container as well as bypassing the VPN for specific websites.

  Read more →
• 

  Hubert Dreyfus's views on artificial intelligence

  Hubert Dreyfus was a critic of artificial intelligence research. In a series of papers and books, including Alchemy and AI (1965), What Computers Can't Do (1972; 1979; 1992) and Mind over Machine (1986), he presented a skeptical and cautious assessment of AI's progress and a critique of the philosophical foundations of the field. Dreyfus' objections are discussed in most introductions to the philosophy of artificial intelligence, including Russell & Norvig (2021), a standard AI textbook, and in Fearn (2007), a survey of contemporary philosophy. Dreyfus argued that human intelligence and expertise depend primarily on yet-to-be understood informal and unconscious processes rather than symbolic manipulation and that these essentially human skills cannot be fully captured in formal rules. His critique was based on the insights of modern continental philosophers such as Merleau-Ponty and Heidegger, and was directed at the first wave of AI research which tried to reduce intelligence to high level formal symbols. When Dreyfus' ideas were first introduced in the mid-1960s, they were met in the AI community with ridicule and outright hostility. By the 1980s, however, some of his perspectives were rediscovered by researchers working in robotics and the new field of connectionism—approaches that were called "sub-symbolic" at the time because they eschewed early AI research's emphasis on high level symbols. In the 21st century, "sub-symbolic" artificial neural networks and other statistics-based approaches to machine learning were highly successful. Historian and AI researcher Daniel Crevier wrote: "time has proven the accuracy and perceptiveness of some of Dreyfus's comments." Dreyfus said in 2007, "I figure I won and it's over—they've given up." == Dreyfus' critique == === The grandiose promises of artificial intelligence === In Alchemy and AI (1965) and What Computers Can't Do (1972), Dreyfus summarized the history of artificial intelligence and ridiculed the unbridled optimism that permeated the field. For example, Herbert A. Simon, following the success of his program General Problem Solver (1957), predicted that by 1967: A computer would be world champion in chess. A computer would discover and prove an important new mathematical theorem. Most theories in psychology will take the form of computer programs. The press dutifully reported these predictions of the imminent arrival of machine intelligence. Dreyfus felt that this optimism was unwarranted and, in 1965, argued forcefully that predictions like these would not come true. He would eventually be proven right. Pamela McCorduck explains Dreyfus' position: A great misunderstanding accounts for public confusion about thinking machines, a misunderstanding perpetrated by the unrealistic claims researchers in AI have been making, claims that thinking machines are already here, or at any rate, just around the corner. These predictions were based on the success of the cognitive revolution, which promoted an "information processing" model of the mind. It was articulated by Newell and Simon in their physical symbol systems hypothesis, and later expanded into a philosophical position known as computationalism by philosophers such as Jerry Fodor and Hilary Putnam. In AI, the approach is now called symbolic AI or "GOFAI". Dreyfus argued that "symbolic AI" was the latest version of the ancient program of rationalism in philosophy. Rationalism had come under heavy criticism in the 20th century from philosophers like Martin Heidegger and Edmund Husserl. The mind, according to modern continental philosophy, is not "rationalist" and is nothing like a digital computer. Cognitivism led early AI researchers to believe that they had successfully simulated the essential process of human thought, thus it seemed a short step to producing fully intelligent machines. Dreyfus' last paper detailed the ongoing history of the "first step fallacy", where AI researchers tend to wildly extrapolate initial success as promising, perhaps even guaranteeing, wild future successes. === Dreyfus' four assumptions of artificial intelligence research === In Alchemy and AI and What Computers Can't Do, Dreyfus identified four philosophical assumptions, at least one of which he deems necessary for AI to succeed. "In each case," Dreyfus writes, "the assumption is taken by workers in AI as an axiom, guaranteeing results, whereas it is, in fact, one hypothesis among others, to be tested by the success of such work." Dreyfus argues that AI would be impossible without accepting at least one of these four assumptions: The biological assumption The brain processes information in discrete operations by way of some biological equivalent of on/off switches. In the early days of research into neurology, scientists found that neurons fire in all-or-nothing pulses. Several researchers, such as Walter Pitts and Warren McCulloch, speculated with great confidence that neurons functioned similarly to the way Boolean logic gates operate, and so could be imitated by electronic circuitry at the level of the neuron. When digital computers became widely used in the early 50s, this argument was extended to suggest that the brain was a vast physical symbol system, manipulating the binary symbols of zero and one. Dreyfus was able to refute the biological assumption by citing research in neurology that suggested that the action and timing of neuron firing had analog components. But Daniel Crevier observes that "few still held that belief in the early 1970s, and nobody argued against Dreyfus" about the biological assumption. The psychological assumption The mind can be viewed as a device operating on bits of information according to formal rules. He refuted this assumption by showing that much of what we know about the world consists of complex attitudes or tendencies that make us lean towards one interpretation over another. He argued that, even when we use explicit symbols, we are using them against an unconscious and informal background including commonsense knowledge and that without this background our symbols cease to mean anything. This background, in Dreyfus' view, was not implemented in individual brains as explicit individual symbols with explicit individual meanings. The epistemological assumption All knowledge can be formalized. This concerns the philosophical issue of epistemology, or the study of knowledge. Even if we agree that the psychological assumption is false, AI researchers could still argue (as AI founder John McCarthy has) that it is possible for a symbol processing machine to represent all knowledge, regardless of whether human beings represent knowledge the same way. Dreyfus argued that there is no justification for this assumption, since so much of human knowledge is not symbolic or even expressible using formal constructs. The ontological assumption The world consists of independent facts that can be represented by independent symbols AI researchers (and futurists and science fiction writers) often assume that there is no limit to formal, scientific knowledge, because they assume that any phenomenon in the universe can be described by symbols or scientific theories. This assumes that everything that exists can be understood as objects, properties of objects, classes of objects, relations of objects, and so on: precisely those things that can be described by logic, language and mathematics. The study of being or existence is called ontology, and so Dreyfus calls this the ontological assumption. If this is false, then it raises doubts about what we can ultimately know and what intelligent machines will ultimately be able to help us to do. === Knowing-how vs. knowing-that: the primacy of intuition === In Mind Over Machine (1986), written (with his brother) during the heyday of expert systems, Dreyfus analyzed the difference between human expertise and the programs that claimed to capture it. This expanded on ideas from What Computers Can't Do, where he had made a similar argument criticizing the "cognitive simulation" school of AI research practiced by Allen Newell and Herbert A. Simon in the 1960s. Dreyfus argued that human problem solving and expertise depend on our background sense of the context, of what is important and interesting given the situation, rather than on the process of searching through combinations of possibilities to find what we need. Dreyfus would describe it in 1986 as the difference between "knowing-that" and "knowing-how", based on Heidegger's distinction of present-at-hand and ready-to-hand. Knowing-that is our conscious, step-by-step problem solving abilities. We use these skills when we encounter a difficult problem that requires us to stop, step back and search through ideas one at time. At moments like this, the ideas become very precise and simple: they become context free symbols, which we manipulate using logic and language. These are the skills that Newell and Simon had demonstrated with both psy

  Read more →
• 

  Trustworthy AI

  Trustworthy AI refers to artificial intelligence systems that are designed to have transparent reasoning, are explainable (XAI), accountable, robust, fair and honest, respectful of data privacy, and steerable or alignable with human goals. == Terminology == Recent work in AI ethics distinguishes trustworthiness and trustability as two different conditions relevant to trustworthy AI. Trustworthiness is concerned with whether an AI system or the institutions deploying it merit trust by being reliable, fair, and accountable. Trustability, on the other hand, is the prior question of whether a given entity is even the kind of thing to which interpersonal trust can coherently apply as opposed to mere instrumental reliance. Some philosophers argue that current AI systems are best understood as tools that are not genuine targets of interpersonal trust. They argue that trust should be directed toward the human and institutional arrangements that govern the systems' design, deployment, and oversight. This stance supports interpreting "trustworthy AI" as trustworthy governance and use of AI rather than trust in the artifacts themselves. Transparency in AI involves making the processes and decisions of such systems understandable to users and stakeholders. Accountability ensures that there are protocols for addressing adverse outcomes or biases that may arise, with designated responsibilities for oversight and remediation. Robustness and security aim to ensure that AI systems perform reliably under various conditions and are safeguarded against malicious attacks. Harmlessness can be achieved by refusal training: training the models to avoid problematic requests, and by adding filters to detect and prevent discussion on biased, unethical, or dangerous outputs. There is research on how to train AI so that it aligns with human goals. == Techniques and ITU standardization == Trustworthy AI creation is a goal of AI governance and policymaking. To achieve transparency and data privacy, several privacy-enhancing technologies (PETs) can be used. These include: Homomorphic encryption for computing with encrypted data without ever decrypting it. Federated learning and secure multi-party computation (MPC) for distributing the model training without sharing information between the learning centers and computing servers. Differential privacy for exposing statistical data while guaranteeing that no private information is exposed. Zero-knowledge proof - providing proven validity for statements without disclosing any extra information. A work programme for achieving Trustworthy AI was set up by the International Telecommunication Union, an agency of the United Nations, initiated under its AI for Good programme. Its origin lies with the ITU-WHO Focus Group on Artificial Intelligence for Health, where a strong need for both privacy and analytics created demand for a standard in these technologies. In 2020, AI for Good moved online, and the TrustworthyAI seminar series was established to initiate discussions on these topics. This eventually led to standardization activities. === Multi-party computation === Secure multi-party computation (MPC) is being standardized under "Question 5" (the incubator) of ITU-T Study Group 17. === Homomorphic encryption === Homomorphic encryption allows for computing on encrypted data, where the outcomes or result is still encrypted and unknown to those performing the computation, but can be deciphered by the original encryptor. It is often developed with the goal of enabling use in jurisdictions different from the data creation (under, for instance, GDPR). ITU has been collaborating since the early stage of the HomomorphicEncryption.org standardization meetings, which has developed a standard on homomorphic encryption. The fifth homomorphic encryption meeting was hosted at ITU HQ in Geneva. === Federated learning === Zero-sum masks as used by federated learning for privacy preservation are used extensively in the multimedia standards of ITU-T Study Group 16 (VCEG) such as JPEG, MP3, H.264, and H.265 (commonly known as MPEG). === Zero-knowledge proof === Previous pre-standardization work on the topic of zero-knowledge proof has been conducted in the ITU-T Focus Group on Digital Ledger Technologies. === Differential privacy === The application of differential privacy in the preservation of privacy was examined at several of the "Day 0" machine learning workshops at AI for Good Global Summits. == Mozilla "Rebel Alliance" == In January 2026, the Mozilla Foundation and its subsidiaries announced a strategic shift to deploy their entire $1.4 billion reserve into building what foundation president Mark Surman termed a "rebel alliance" for trustworthy AI. Framed by Surman as a mission-driven alternative to the market dominance of OpenAI and Anthropic, the initiative seeks to establish an open-source AI stack by 2028. The alliance includes several startups funded via Mozilla Ventures, specifically focusing on decentralized governance and transparency: Trail: A firm developing AI compliance frameworks for regulated industries. Transformer Lab: A developer of open-source tools for AI model management. Oumi: A platform for training and deploying open-source models. The "rebel alliance" terminology is a historical reference to Mozilla's efforts in 1998 to challenge Microsoft's browser monopoly. While the $1.4 billion in funding is significant, it has been contrasted with the tens of billions in capital raised by proprietary competitors like OpenAI.

  Read more →
• 

  Leela Zero

  Leela Zero is a free and open-source computer Go program released on 25 October 2017. It is developed by Belgian programmer Gian-Carlo Pascutto, the author of chess engine Sjeng and Go engine Leela. Leela Zero's algorithm is based on DeepMind's 2017 paper about AlphaGo Zero. Unlike the original Leela, which has a lot of human knowledge and heuristics programmed into it, the program code in Leela Zero only knows the basic rules and nothing more. The knowledge that makes Leela Zero a strong player is contained in a neural network, which is trained based on the results of previous games that the program played. Leela Zero is trained by a distributed effort, which is coordinated at the Leela Zero website. Members of the community provide computing resources by running the client, which generates self-play games and submits them to the server. The self-play games are used to train newer networks. Generally, over 500 clients have connected to the server to contribute resources. The community has provided high quality code contributions as well. == Version history == Leela Zero finished third at the BerryGenomics Cup World AI Go Tournament in Fuzhou, Fujian, China on 28 April 2018. The New Yorker at the end of 2018 characterized Leela and Leela Zero as "the world’s most successful open-source Go engines". In early 2018, another team branched Leela Chess Zero from the same code base, also to verify the methods in the AlphaZero paper as applied to the game of chess. AlphaZero's use of Google TPUs was replaced by a crowd-sourcing infrastructure and the ability to use graphics card GPUs via the OpenCL library. Even so, it is expected to take a year of crowd-sourced training to make up for the dozen hours that AlphaZero was allowed to train for its chess match in the paper. The distributed training server was shut down on 2021-02-15, marking the end of Leela Zero project. The page now directs visitors to KataGo and SAI. The model sizes increased steadily over time. The first released model has hash name d645af97, size 1x8 (1 layer, 8 channels), and released at 2017-11-10 13:04. The last released model has hash name 0e9ea880, size 40x256, and was released at 2021-02-15 09:04. == Technology == Leela Zero is an (almost) exact replication of AlphaGo Zero in both training process and architecture. The training process is Monte-Carlo Tree Search with self-play, exactly the same as AlphaGo Zero. The architecture is the same as AlphaGo Zero (with one difference). Consider the last released model, 0e9ea880. It has 47 million parameters, and the following architecture: The stem of the network takes as input a 18x19x19 tensor representation of the Go board. 8 channels are the positions of the current player's stones from the last eight time steps. (1 if there is a stone, 0 otherwise. If the time step go before the beginning of the game, then 0 in all positions.) 8 channels are the positions of the other player's stones from the last eight time steps. 1 channel is all 1 if black is to move, and 0 otherwise. 1 channel is all 1 if white is to move, and 0 otherwise. (This channel is not present in the original AlphaGo Zero) The body is a ResNet with 40 residual blocks and 256 channels. There are two heads, a policy head and a value head. Policy head outputs a logit array of size 19 × 19 + 1 {\displaystyle 19\times 19+1} , representing the logit of making a move in one of the points, plus the logit of passing. Value head outputs a number in the range ( − 1 , + 1 ) {\displaystyle (-1,+1)} , representing the expected score for the current player. -1 represents current player losing, and +1 winning.

  Read more →
• 

  Knowledge graph embedding

  In representation learning, knowledge graph embedding (KGE), also called knowledge representation learning (KRL), or multi-relation learning, is a machine learning task of learning a low-dimensional representation of a knowledge graph's entities and relations while preserving their semantic meaning. Leveraging their embedded representation, knowledge graphs can be used for various applications such as link prediction, triple classification, entity recognition, clustering, and relation extraction. == Definition == A knowledge graph G = { E , R , F } {\displaystyle {\mathcal {G}}=\{E,R,F\}} is a collection of entities E {\displaystyle E} , relations R {\displaystyle R} , and facts F {\displaystyle F} . A fact is a triple ( h , r , t ) ∈ F {\displaystyle (h,r,t)\in F} that denotes a link r ∈ R {\displaystyle r\in R} between the head h ∈ E {\displaystyle h\in E} and the tail t ∈ E {\displaystyle t\in E} of the triple. Another notation that is often used in the literature to represent a triple (or fact) is ⟨ head , relation , tail ⟩ {\displaystyle \langle {\text{head}},{\text{relation}},{\text{tail}}\rangle } . This notation is called the Resource Description Framework (RDF). A knowledge graph represents the knowledge related to a specific domain; leveraging this structured representation, it is possible to infer a piece of new knowledge from it after some refinement steps. However, nowadays, people have to deal with the sparsity of data and the computational inefficiency to use them in a real-world application. The embedding of a knowledge graph is a function that translates each entity and each relation into a vector of a given dimension d {\displaystyle d} , called embedding dimension. It is even possible to embed the entities and relations with different dimensions. The embedding vectors can then be used for other tasks. A knowledge graph embedding is characterized by four aspects: Representation space: The low-dimensional space in which the entities and relations are represented. Scoring function: A measure of the goodness of a triple-embedded representation. Encoding models: The modality in which the embedded representation of the entities and relations interact with each other. Additional information: Any additional information coming from the knowledge graph that can enrich the embedded representation. Usually, an ad hoc scoring function is integrated into the general scoring function for each additional piece of information. == Embedding procedure == All algorithms for creating a knowledge graph embedding follow the same approach. First, the embedding vectors are initialized to random values. Then, they are iteratively optimized using a training set of triples. In each iteration, a batch of size b {\displaystyle b} triples is sampled from the training set, and a triple from it is sampled and corrupted—i.e., a triple that does not represent a true fact in the knowledge graph. The corruption of a triple involves substituting the head or the tail (or both) of the triple with another entity that makes the fact false. The original triple and the corrupted triple are added in the training batch, and then the embeddings are updated, optimizing a scoring function. Iteration stops when a stop condition is reached. Usually, the stop condition depends on the overfitting of the training set. At the end, the learned embeddings should have extracted semantic meaning from the training triples and should correctly predict unseen true facts in the knowledge graph. === Pseudocode === The following is the pseudocode for the general embedding procedure. algorithm Compute entity and relation embeddings input: The training set S = { ( h , r , t ) } {\displaystyle S=\{(h,r,t)\}} , entity set E {\displaystyle E} , relation set R {\displaystyle R} , embedding dimension k {\displaystyle k} output: Entity and relation embeddings initialization: the entities e {\displaystyle e} and relations r {\displaystyle r} embeddings (vectors) are randomly initialized while stop condition do S b a t c h ← s a m p l e ( S , b ) {\displaystyle S_{batch}\leftarrow sample(S,b)} // Sample a batch from the training set for each ( h , r , t ) {\displaystyle (h,r,t)} in S b a t c h {\displaystyle S_{batch}} do ( h ′ , r , t ′ ) ← s a m p l e ( S ′ ) {\displaystyle (h',r,t')\leftarrow sample(S')} // Sample a corrupted fact T b a t c h ← T b a t c h ∪ { ( ( h , r , t ) , ( h ′ , r , t ′ ) ) } {\displaystyle T_{batch}\leftarrow T_{batch}\cup \{((h,r,t),(h',r,t'))\}} end for Update embeddings by minimizing the loss function end while == Performance indicators == These indexes are often used to measure the embedding quality of a model. The simplicity of the indexes makes them very suitable for evaluating the performance of an embedding algorithm even on a large scale. Given Q {\displaystyle {\ce {Q}}} as the set of all ranked predictions of a model, it is possible to define three different performance indexes: Hits@K, MR, and MRR. === Hits@K === Hits@K or in short, H@K, is a performance index that measures the probability to find the correct prediction in the first top K model predictions. Usually, it is used k = 10 {\displaystyle k=10} . Hits@K reflects the accuracy of an embedding model to predict the relation between two given triples correctly. Hits@K = | { q ∈ Q : q < k } | | Q | ∈ [ 0 , 1 ] {\displaystyle ={\frac {|\{q\in Q:q Read more →

• 

  AGROVOC

  AGROVOC is a multilingual controlled vocabulary covering areas of interest of the Food and Agriculture Organization of the United Nations (FAO), aiming to promote the visibility of research produced among FAO members. By March 2024, AGROVOC consisted of over 42 000 concepts and up to 1 000 000 terms in more than 42 different languages. It is a collaborative effort, the outcome of consensus among a community of experts coordinated by FAO. == History == FAO first published AGROVOC at the beginning of the 1980s in English, Spanish and French to serve as a controlled vocabulary to index publications in agricultural science and technology, especially for the International System for Agricultural Science and Technology (AGRIS). In the 1990s, AGROVOC shifted from paper printing to a digital format opting for data storage handled by a relational database. In 2004, preliminary experiments with expressing AGROVOC into the Web Ontology Language (OWL) took place. At the same time a web based editing tool was developed, then called WorkBench, nowadays VocBench. In 2009 AGROVOC became an SKOS resource. == Usage == Today, AGROVOC is available in different languages. It is employed for tagging resources, allowing searches in a specific language while providing results in many others, enhancing their visibility worldwide. Additionally, it serves for organizing knowledge to facilitate subsequent data retrieval, tagging website content for search engine discovery, standardizing agricultural information data and acting as a reference for translations. Moreover, it finds applications in fields such as data mining, big data, or artificial intelligence. Updated AGROVOC content is released once a month and is available for public use. == Maintenance == FAO coordinates the editorial activities related to the maintenance of AGROVOC. Content curation is carried out by a community of editors and institutions responsible for each of the language versions. VocBench, is the tool used to edit and maintain AGROVOC in a distributed way. FAO also facilitates the technical maintenance of AGROVOC. == Copyright and license == Copyright for AGROVOC content in FAO languages (English, French, Spanish, Arabic, Russian and Chinese) is held by FAO, while content in other languages stays with the institutions that authored it. AGROVOC thesaurus content in English, Russian, French, Spanish, Arabic and Chinese is licensed under the international Creative Commons Attribution License (CC-BY-4.0).

  Read more →
• 

  Yale shooting problem

  The Yale shooting problem is a conundrum or scenario in formal situational logic on which early logical solutions to the frame problem fail. The name of this problem comes from a scenario proposed by its inventors, Steve Hanks and Drew McDermott, working at Yale University when they proposed it. In this scenario, Fred (later identified as a turkey) is initially alive and a gun is initially unloaded. Loading the gun, waiting for a moment, and then shooting the gun at Fred is expected to kill Fred. However, if inertia is formalized in logic by minimizing the changes in this situation, then it cannot be uniquely proved that Fred is dead after loading, waiting, and shooting. In one solution, Fred indeed dies; in another (also logically correct) solution, the gun becomes mysteriously unloaded and Fred survives. Technically, this scenario is described by two fluents (a fluent is a condition that can change truth value over time): a l i v e {\displaystyle alive} and l o a d e d {\displaystyle loaded} . Initially, the first condition is true and the second is false. Then, the gun is loaded, some time passes, and the gun is fired. Such problems can be formalized in logic by considering four time points 0 {\displaystyle 0} , 1 {\displaystyle 1} , 2 {\displaystyle 2} , and 3 {\displaystyle 3} , and turning every fluent such as a l i v e {\displaystyle alive} into a predicate a l i v e ( t ) {\displaystyle alive(t)} depending on time. A direct formalization of the statement of the Yale shooting problem in logic is the following one: a l i v e ( 0 ) {\displaystyle alive(0)} ¬ l o a d e d ( 0 ) {\displaystyle \neg loaded(0)} t r u e → l o a d e d ( 1 ) {\displaystyle true\rightarrow loaded(1)} l o a d e d ( 2 ) → ¬ a l i v e ( 3 ) {\displaystyle loaded(2)\rightarrow \neg alive(3)} The first two formulae represent the initial state. The third formula formalizes the effect of loading the gun at time 1 {\displaystyle 1} . The fourth formula formalizes the effect of shooting at Fred at time 2 {\displaystyle 2} . This is a simplified formalization in which action names are neglected and the effects of actions are directly specified for the time points in which the actions are executed. See situation calculus for details. The formulae above, while being direct formalizations of the known facts, do not suffice to correctly characterize the domain. Indeed, ¬ a l i v e ( 1 ) {\displaystyle \neg alive(1)} is consistent with all these formulae, although there is no reason to believe that Fred dies before the gun has been shot. The problem is that the formulae above only include the effects of actions, but do not specify that all fluents not changed by the actions remain the same. In other words, a formula a l i v e ( 0 ) ≡ a l i v e ( 1 ) {\displaystyle alive(0)\equiv alive(1)} must be added to formalize the implicit assumption that loading the gun only changes the value of l o a d e d {\displaystyle loaded} and not the value of a l i v e {\displaystyle alive} . The necessity of a large number of formulae stating the obvious fact that conditions do not change unless an action changes them is known as the frame problem. An early solution to the frame problem was based on minimizing the changes. In other words, the scenario is formalized by the formulae above (that specify only the effects of actions) and by the assumption that the changes in the fluents over time are as minimal as possible. The rationale is that the formulae above enforce all effect of actions to take place, while minimization should restrict the changes to exactly those due to the actions. In the Yale shooting scenario, one possible evaluation of the fluents in which the changes are minimized is the following one. This is the expected solution. It contains two fluent changes: l o a d e d {\displaystyle loaded} becomes true at time 1 and a l i v e {\displaystyle alive} becomes false at time 3. The following evaluation also satisfies all formulae above. In this evaluation, there are still two changes only: l o a d e d {\displaystyle loaded} becomes true at time 1 and false at time 2. As a result, this evaluation is considered a valid description of the evolution of the state, although there is no valid reason to explain l o a d e d {\displaystyle loaded} being false at time 2. The fact that minimization of changes leads to wrong solution is the motivation for the introduction of the Yale shooting problem. While the Yale shooting problem has been considered a severe obstacle to the use of logic for formalizing dynamical scenarios, solutions to it have been known since the late 1980s. One solution involves the use of predicate completion in the specification of actions: in this solution, the fact that shooting causes Fred to die is formalized by the preconditions: alive and loaded, and the effect is that alive changes value (since alive was true before, this corresponds to alive becoming false). By turning this implication into an if and only if statement, the effects of shooting are correctly formalized. (Predicate completion is more complicated when there is more than one implication involved.) A solution proposed by Erik Sandewall was to include a new condition of occlusion, which formalizes the “permission to change” for a fluent. The effect of an action that might change a fluent is therefore that the fluent has the new value, and that the occlusion is made (temporarily) true. What is minimized is not the set of changes, but the set of occlusions being true. Another constraint specifying that no fluent changes unless occlusion is true completes this solution. The Yale shooting scenario is also correctly formalized by the Reiter version of the situation calculus, the fluent calculus, and the action description languages. In 2005, the 1985 paper in which the Yale shooting scenario was first described received the AAAI Classic Paper award. In spite of being a solved problem, that example is still sometimes mentioned in recent research papers, where it is used as an illustrative example (e.g., for explaining the syntax of a new logic for reasoning about actions), rather than being presented as a problem.

  Read more →