Reciprocal human machine learning

Reciprocal Human Machine Learning (RHML) is an interdisciplinary approach to designing human-AI interaction systems. RHML aims to enable continual learning between humans and machine learning models by having them learn from each other. This approach keeps the human expert "in the loop" to oversee and enhance machine learning performance and simultaneously support the human expert continue learning. == Background == RHML emerged in the context of the rise of big data analytics and artificial intelligence for intelligent tasks like sense-making and decision-making. As machine learning advanced to take on more roles, researchers realized fully autonomous systems had limitations and needed human guidance. RHML extends the concept of human-in-the-loop systems by promoting reciprocal learning. Humans learn from their interactions with machine learning models, staying up-to-date on evolving technology. The models also learn from human feedback and oversight. This amplification of learning on both sides is a key focus of RHML. The approach draws on theories of learning in dyads from education and psychology. It also builds on human-computer interaction and human-centered design principles. Implementing RHML requires developing specialized tools and interfaces tailored to the application == Applications == RHML has been explored across diverse domains including: Cybersecurity - Software to enable reciprocal learning between experts and AI models for social media threat detection. Organizational decision-making - RHML to structure collaboration between humans and AI systems. Workplace training - Using RHML for workers to learn from AI technologies on the job. Open science - Using human and AI collaboration to promote open science. Production and logistics - turning workers and intelligent machines into teammates. RHML maintains human oversight and control over AI systems, while enabling cutting-edge machine learning performance. This collaborative approach highlights the importance of keeping the human expert involved in the loop. An example of RHML in application is Free Spirit (AFSFCV), an open-source architecture first published in early 2025 as a whitepaper, proposing a visually structured approach to intent-based human–AI interaction.

List of publications in data science

This is a list of publications in data science, generally organized by order of use in a data analysis workflow. See the list of publications in statistics for more research-based and fundamental publications; while this list is more applied, business oriented, and cross-disciplinary. General article inclusion criteria are: Papers from notable practitioners or notable professors, either with a Wikipedia page or reference to their notability Common knowledge all data professionals should know, with references validating this claim Highly cited applied statistics and machine learning publications Discussion-facilitating papers on the field of data science as a whole (for example, the Attention Is All You Need paper is arguably a landmark paper that can be added here, but it is specific to generative artificial intelligence, not for all practitioners of data) Some reasons why a particular publication might be regarded as important: Topic creator – A publication that created a new topic Breakthrough – A publication that changed scientific knowledge significantly Influence – A publication which has significantly influenced the world or has had a massive impact on the teaching of data science. When possible, a reference is used to validate the inclusion of the publication in this list. == History == Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author) Author: Leo Breiman Publication data: Online version: https://projecteuclid.org/journals/statistical-science/volume-16/issue-3/Statistical-Modeling--The-Two-Cultures-with-comments-and-a/10.1214/ss/1009213726.pdf Description: Describes two cultures of statistics, one using a parsimonious and generative stochastic model, while the other is an algorithmic model with no known mechanism for how the data is generated. Breiman argues that while statistics has traditionally favored using the stochastic model, there is value in expanding the methods that statisticians can use to study phenomenon. Importance: Influence on the philosophies of statisticians right before the increased use of machine learning and deep learning methods. In a 20-year retrospective on this article, "Breiman's words are perhaps more relevant than ever". Notable statisticians at the time wrote opinion pieces about the publication. Although overall critical of the publication, David Cox writes that the publication "contains enough truth and exposes enough weaknesses to be thought-provoking." Bradley Efron commented that this publication is a "stimulating paper". Emanuel Parzen also comments about this publication that "Breiman alerts us to systematic blunders (leading to wrong conclusions) that have been committed applying current statistical practice of data modeling". Data Scientist: The Sexiest Job of the 21st Century Author: Thomas H. Davenport and DJ Patil Publication data: Online version: hbr.org/2022/07/is-data-scientist-still-the-sexiest-job-of-the-21st-century Description: Describes the new role at companies that is coined "Data scientist", what they do, how an organization might recruit one to their organization, and how to work with one effectively. Importance: This publication has been an influence on the data community as mentioned near the time it was published in 2012 by institutions like IEEE Spectrum, but also mentioned nearly a decade later asking the same question the title poses. In a retrospective response to their own publication 10 years earlier, authors Davenport and Patil have reflected that the role of a data scientist has "become better institutionalized, the scope of the job has been redefined, the technology it relies on has made huge strides, and the importance of non-technical expertise, such as ethics and change management, has grown". 50 Years of Data Science Author: David Donoho Publication data: Online version: https://www.tandfonline.com/doi/full/10.1080/10618600.2017.1384734 Description: Retrospective discussion paper on the history and origins of data science, with a number of commentary from notable statisticians. Importance: This has been described as "the first in the field to present such a comprehensive and in-depth survey and overview", and helps to define the field that has many definitions. The Composable Data Management System Manifesto Author: Pedro Pedreira, Orri Erling, Konstantinos Karanasos, Scott Schneider, Wes McKinney, Satya R Valluri, Mohamed Zait, Jacques Nadeau Publication data: Online version: https://www.vldb.org/pvldb/vol16/p2679-pedreira.pdf Description: The vision paper advocating for a paradigm shift in how data management systems are designed using standard, composable, interoperable tools rather than siloed software tools. Importance: A paradigm shifting view on how future data science software tools should be designed for more efficient workflows, the principles of which "will be especially crucial for addressing fragmentation, improving interoperability, and promoting user-centricity as data ecosystems grow increasingly complex". == Data collection and organization == Tidy Data Author: Hadley Wickham Publication data: Online version: https://www.jstatsoft.org/article/view/v059i10/ https://vita.had.co.nz/papers/tidy-data.pdf Description: Describes a framework for data cleaning that is summarized in the quote, "each variable is a column, each observation is a row, and each type of observational unit is a table". This allows a standard data structure for which data analysis tools can be consistently built around. Importance: Cited over 1,500 times, this effort for tidy data has been described by David Donoho as having "more impact on today's practice of data analysis than many highly regarded theoretical statistics articles". In the context of data visualization, this publication is said to support "efficient exploration and prototyping because variables can be assigned different roles in the plot without modifying anything about the original dataset". Data Organization in Spreadsheets Author: Karl W. Broman and Kara H. Woo Publication data: Online version: https://www.tandfonline.com/doi/full/10.1080/00031305.2017.1375989 Description: This article offers practical recommendations for organizing data in spreadsheets, like Microsoft Excel and Google Sheets, to reduce errors and lower the barrier for later analyses due to limitations in spreadsheets or quirks in the software. Importance: Influences teaching both data and non-data practitioners to create more analysis-friendly spreadsheets, and has been described to outline "spreadsheet best practices". == Data visualizations == Quantitative Graphics in Statistics: A Brief History Author: James R. Beniger and Dorothy L. Robyn Publication data: Online version: https://www.jstor.org/stable/2683467 Description: Outlines history and evolution of quantitative graphics in statistics, going through spatial organization (17th and 18th centuries), discrete comparison (18th and 19th centuries), continuous distribution (19th century), and multivariate distribution and correlation (late 19th and 20th centuries). Importance: Helps put into perspective for learning data practitioners the recency of graphics that are used. A later publication "Graphical Methods in Statistics" by Stephen Fienberg in 1979 writes that his publication "owes much to the work of Beniger and Robyn". == Practice == Data Science for Business Author: Foster Provost and Tom Fawcett Publication data: Online version: N/A Description: Broadly outlines principles of data science and data-analytic thinking for businesses. Importance: Cited over 3,000 times, it is "highly recommended for students" but also it is also recommended due to its "relevance to senior management leaders who want to build and lead a team of data scientists and implement data science in solving complex business problems". == Tooling == Hidden Technical Debt in Machine Learning Systems Author: D. Sculley, Gary Holy, Daniel Golovin, Eugene Davydov, Todd Phillips, Dietmar Ebner, Vinay Chaudhary, Michael Young, Jean-François Crespo, Dan Dennison Publication data: Online version: https://proceedings.neurips.cc/paper_files/paper/2015/file/86df7dcfd896fcaf2674f757a2463eba-Paper.pdf Description: This paper argues that it is "dangerous to think of [complex machine learning] quick wins as coming for free" and overviews risk factors to account for when implementing a machine learning system. Importance: All authors worked for Google, article is cited over 2,000 times, and helped practitioners thinking about quickly implementing a machine learning tool without understanding the long-term maintenance of the tool. A few useful things to know about machine learning Author: Pedro Domingos Publication data: Online version: https://dl.acm.org/doi/10.1145/2347736.2347755 https://homes.cs.washington.edu/~pedrod/papers/cacm12.pdf Description: The purpose of this paper is to distill inaccessible "folk knowledge" to effectively implement machine learning projects because "machin

Timeline of artificial intelligence risks in global finance

The following article is a broad timeline of the course of events related to artificial intelligence risks in global finance. The AI boom has led to concerns including the existential risk from artificial intelligence, as the uptake on applications of artificial intelligence increases. By late 2025, global finance and artificial intelligence were "deeply intertwined". A June 2025 Menlo Ventures report raised concerns about the sustainability of future revenue and long-term profitability of AI, given the relatively low rate of consumer monetization. == 2017 == 30 NovemberThe New York Times said that new AI reports by McKinsey & Company, the National Bureau of Economic Research, and an AI Index created by university researchers, indicated an early AI boom. The Index built on a project—"The One Hundred Year Study on Artificial Intelligence" launched in 2014. == 2018 == 2018 was a year of incremental AI growth in finance. == 2022 == The release of ChatGPT by OpenAI became the catalyst for an artificial intelligence boom that continues to remake the global economy. According to a European Central Bank report, public interest in AI increased rapidly as evidenced with rising Google searches, AI jobs, models, patents, and innovations since late 2022. At that time Europe led the US in the size of its AI workforce. == 2023 == The regulatory body, the International Monetary Fund (IMF), published their report, "Generative Artificial Intelligence in Finance: Risk Considerations", drawing attention to oversight gaps and the need for regulations. The report explores the risks posed by using generative artificial intelligence (GenAI) systems in the financial sector including "broader risks to financial stability." == 2024 == January 12 In January 2024 Bloomberg's published its list of the "Magnificent Seven" Big Tech companies on the stock market based on their strength, size and market capitalization:Apple, Microsoft, Alphabet (Google), Amazon, Meta Platforms (Facebook), Nvidia, and Tesla. 21 June During the AI boom, Nvidia became the world's most valuable company, surpassing Microsoft, as its value increased to over US$4 trillion. In 2023 and 2024, the "Magnificent Seven" stocks were the primary drivers behind the increase in equity indexes, according to Reuters. == 2025 == === January === 23 January President Donald Trump's AI policy was announced calling for United States global leadership in artificial intelligence. The Economist noted that this politic shift in which the United States seeks "global dominance" in AI includes trimming regulations and assisting in expansion of infrastructure and increase in number of AI workers. Governments of Gulf nations were also investing trillions of dollars in AI. 27 January Against the backdrop of a tech war between China and the United States over AI dominance, within days of the launch of China's free DeepSeek App, it was the most downloaded app in the United States, rising to the first place in the Apple app store. President Trump responded immediately, saying this "sudden rise" should be a "wake-up" call to the United States, and called on US companies to be more competitive. === June === 26 June In their June 2025 report, Menlo Ventures estimated that only about 3% of consumers paid for artificial intelligence-related services, representing about $USD12 billion in annual spending. This is relatively low in contrast to the massive capital expenditure by AI infrastructure companies, which raises concerns about revenue sustainability and long-term profitability. === July === 23 July The Trump administration launched the US AI Action Plan, positioning the United States in a high-stakes technological race with China for global dominance in artificial intelligence, emphasizing that neither nation can afford to fall behind due to the exponential nature of AI advancement. The plan, a new government website and policy speech called for accelerated AI adoption across federal agencies, and a number of initiatives to make is easier for AI infrastructure expansion, and other measures to ensure American leadership in AI standards. Some leading experts warned that the administration failed to provide sufficient regulations and safeguards for AI safety. Concerns were raised about the negative impacts of cuts to research funding and tightened visa policies for scientists, potentially undermining public trust and America's ability to compete internationally. === September === 7 September The Economist cautioned that AI revenues are relatively modest compared to the high cost and investments in the creation of new data centers. Even Sam Altman, OpenAI CEO and one of the leading figures of the AI boom,, raised concerns about investors' outsized hopes for financial returns. At the same time, history has shown that new technologies, like railways and electricity, endured and spread after the initial hype faded. 12 September Economists warn that U.S. households' direct and indirect investments—mutual funds or retirement plans—in the stock market reached an unprecedented historically high level, now representing 45% of all financial assets, or about $USD51.2 trillion. Compared to the Dot-com bubble this represents a sharp increase in exposure. This makes U.S. households vulnerable to market downturns which in turn would result in decreasing consumer spending. U.S. household net worth rose to a record $176.3 trillion in the second quarter, an increase of $7.3 trillion since early 2025 and about $46 trillion higher than before the pandemic. Federal Reserve data attribute the surge primarily to gains in stock markets and housing values. However, the rise in wealth on paper coincided with increased household borrowing and growing government debt. 18 September Questions were being raised about how quickly the data centers, chips, servers, and GPUs assets of major AI companies will depreciate in value. Comparisons have been made to the Railway Mania in the aftermath of the stock market bubble where a valuable physical infrastructure remained standing, and the telecoms crash after the dot-com bubble which left fiber networks. 28 September There were warnings that record-high American stock ownership during the AI-fueled market boom is a red flag for systemic risk, as the current concentration in equities exceeds levels seen before the dot-com bubble burst in 2000, and could amplify the impact of any future stock market correction. === October === 3 October In 2025 alone, venture capitalists invested almost $USD200 billion in the artificial intelligence sector. 29 October Nvidia was the first company in the world to be valued at US$5 trillion, largely due to AI demand and strategic partnerships with leading technology and AI firms. Nvidia's increase in value was "meteoric". === November === 2 November Forbes reported that, since April, the 'Magnificent Seven' tech giants together contributed over 40% of the S&P 500's return, highlighting their outsized influence and the growing impact of AI on market valuations. CNN warned that while there is a current benefit to investors, with such a high concentration in the S&P 500, they are highly exposed to the fate of the Mag Seven. 2 November Globally there are 11,000 datacentres—huge campuses for AI infrastructure, including thousands of chips, GPUS, and servers. This represents a 500% increase over the last two decades. It is anticipated that $3USDtn more will be spent on increasing that number over the next two or three years. 5 November Concerns about the potential for a market bubble were raised as six of the AI-related Big Tech "Magnificent Seven"—that contribute to the AI boom—reported losing ground in the stock market. Global markets and artificial intelligence have become "deeply intertwined", according to a Reuters report. As of November 2025, more than 50% of the 20 largest S&P firms were deeply exposed to AI. In contrast, in 2000, the 20 S&P 500 firms represented 39% of its total value only 11 of these companies were exposed to the internet. If AI fails to deliver strong returns on their investments, these top S&P firms would be significantly impacted, according to the Economist. Analysts suggest that the AI market in 2025 may not behave like a traditional one, as investors are simultaneously aware of the risks and driven by the potential for outsized rewards. Leading AI labs may believe that the first company to achieve artificial general intelligence (AGI), when an AI system surpasses all human cognitive abilities and becomes capable of self-improvement—could dominate the future of technology and finance. While some have estimated that the potential value of such a breakthrough could be as high as $1.46 quadrillion, this figure is speculative and widely debated. 5 November Bloomberg described Nvidia's H100 Hopper-Blackwell AI chips as the "King of AI chips". Nvidia dominates the AI chip market with over 78% of the market share because of both speed and cost. According to B

Empirical dynamic modeling

Empirical dynamic modeling (EDM) is a framework for analysis and prediction of nonlinear dynamical systems. Applications include population dynamics, ecosystem service, medicine, neuroscience, dynamical systems, geophysics, and human-computer interaction. EDM was originally developed by Robert May and George Sugihara. It can be considered a methodology for data modeling, predictive analytics, dynamical system analysis, machine learning and time series analysis. == Description == Mathematical models have tremendous power to describe observations of real-world systems. They are routinely used to test hypothesis, explain mechanisms and predict future outcomes. However, real-world systems are often nonlinear and multidimensional, in some instances rendering explicit equation-based modeling problematic. Empirical models, which infer patterns and associations from the data instead of using hypothesized equations, represent a natural and flexible framework for modeling complex dynamics. Donald DeAngelis and Simeon Yurek illustrated that canonical statistical models are ill-posed when applied to nonlinear dynamical systems. A hallmark of nonlinear dynamics is state-dependence: system states are related to previous states governing transition from one state to another. EDM operates in this space, the multidimensional state-space of system dynamics rather than on one-dimensional observational time series. EDM does not presume relationships among states, for example, a functional dependence, but projects future states from localised, neighboring states. EDM is thus a state-space, nearest-neighbors paradigm where system dynamics are inferred from states derived from observational time series. This provides a model-free representation of the system naturally encompassing nonlinear dynamics. A cornerstone of EDM is recognition that time series observed from a dynamical system can be transformed into higher-dimensional state-spaces by time-delay embedding with Takens's theorem. The state-space models are evaluated based on in-sample fidelity to observations, conventionally with Pearson correlation between predictions and observations. == Methods == Primary EDM algorithms include Simplex projection, Sequential locally weighted global linear maps (S-Map) projection, Multivariate embedding in Simplex or S-Map, Convergent cross mapping (CCM), and Multiview Embeding, described below. Nearest neighbors are found according to: NN ( y , X , k ) = ‖ X N i E − y ‖ ≤ ‖ X N j E − y ‖ if 1 ≤ i ≤ j ≤ k {\displaystyle {\text{NN}}(y,X,k)=\|X_{N_{i}}^{E}-y\|\leq \|X_{N_{j}}^{E}-y\|{\text{ if }}1\leq i\leq j\leq k} === Simplex === Simplex projection is a nearest neighbor projection. It locates the k {\displaystyle k} nearest neighbors to the location in the state-space from which a prediction is desired. To minimize the number of free parameters k {\displaystyle k} is typically set to E + 1 {\displaystyle E+1} defining an E + 1 {\displaystyle E+1} dimensional simplex in the state-space. The prediction is computed as the average of the weighted phase-space simplex projected T p {\displaystyle Tp} points ahead. Each neighbor is weighted proportional to their distance to the projection origin vector in the state-space. Find k {\displaystyle k} nearest neighbor: N k ← NN ( y , X , k ) {\displaystyle N_{k}\gets {\text{NN}}(y,X,k)} Define the distance scale: d ← ‖ X N 1 E − y ‖ {\displaystyle d\gets \|X_{N_{1}}^{E}-y\|} Compute weights: For{ i = 1 , … , k {\displaystyle i=1,\dots ,k} } : w i ← exp ⁡ ( − ‖ X N i E − y ‖ / d ) {\displaystyle w_{i}\gets \exp(-\|X_{N_{i}}^{E}-y\|/d)} Average of state-space simplex: y ^ ← ∑ i = 1 k ( w i X N i + T p ) / ∑ i = 1 k w i {\displaystyle {\hat {y}}\gets \sum _{i=1}^{k}\left(w_{i}X_{N_{i}+T_{p}}\right)/\sum _{i=1}^{k}w_{i}} === S-Map === S-Map extends the state-space prediction in Simplex from an average of the E + 1 {\displaystyle E+1} nearest neighbors to a linear regression fit to all neighbors, but localised with an exponential decay kernel. The exponential localisation function is F ( θ ) = exp ( − θ d / D ) {\displaystyle F(\theta )={\text{exp}}(-\theta d/D)} , where d {\displaystyle d} is the neighbor distance and D {\displaystyle D} the mean distance. In this way, depending on the value of θ {\displaystyle \theta } , neighbors close to the prediction origin point have a higher weight than those further from it, such that a local linear approximation to the nonlinear system is reasonable. This localisation ability allows one to identify an optimal local scale, in-effect quantifying the degree of state dependence, and hence nonlinearity of the system. Another feature of S-Map is that for a properly fit model, the regression coefficients between variables have been shown to approximate the gradient (directional derivative) of variables along the manifold. These Jacobians represent the time-varying interaction strengths between system variables. Find k {\displaystyle k} nearest neighbor: N ← NN ( y , X , k ) {\displaystyle N\gets {\text{NN}}(y,X,k)} Sum of distances: D ← 1 k ∑ i = 1 k ‖ X N i E − y ‖ {\displaystyle D\gets {\frac {1}{k}}\sum _{i=1}^{k}\|X_{N_{i}}^{E}-y\|} Compute weights: For{ i = 1 , … , k {\displaystyle i=1,\dots ,k} } : w i ← exp ⁡ ( − θ ‖ X N i E − y ‖ / D ) {\displaystyle w_{i}\gets \exp(-\theta \|X_{N_{i}}^{E}-y\|/D)} Reweighting matrix: W ← diag ( w i ) {\displaystyle W\gets {\text{diag}}(w_{i})} Design matrix: A ← [ 1 X N 1 X N 1 − 1 … X N 1 − E + 1 1 X N 2 X N 2 − 1 … X N 2 − E + 1 ⋮ ⋮ ⋮ ⋱ ⋮ 1 X N k X N k − 1 … X N k − E + 1 ] {\displaystyle A\gets {\begin{bmatrix}1&X_{N_{1}}&X_{N_{1}-1}&\dots &X_{N_{1}-E+1}\\1&X_{N_{2}}&X_{N_{2}-1}&\dots &X_{N_{2}-E+1}\\\vdots &\vdots &\vdots &\ddots &\vdots \\1&X_{N_{k}}&X_{N_{k}-1}&\dots &X_{N_{k}-E+1}\end{bmatrix}}} Weighted design matrix: A ← W A {\displaystyle A\gets WA} Response vector at T p {\displaystyle Tp} : b ← [ X N 1 + T p X N 2 + T p ⋮ X N k + T p ] {\displaystyle b\gets {\begin{bmatrix}X_{N_{1}+T_{p}}\\X_{N_{2}+T_{p}}\\\vdots \\X_{N_{k}+T_{p}}\end{bmatrix}}} Weighted response vector: b ← W b {\displaystyle b\gets Wb} Least squares solution (SVD): c ^ ← argmin c ‖ A c − b ‖ 2 2 {\displaystyle {\hat {c}}\gets {\text{argmin}}_{c}\|Ac-b\|_{2}^{2}} Local linear model c ^ {\displaystyle {\hat {c}}} is prediction: y ^ ← c ^ 0 + ∑ i = 1 E c ^ i y i {\displaystyle {\hat {y}}\gets {\hat {c}}_{0}+\sum _{i=1}^{E}{\hat {c}}_{i}y_{i}} === Multivariate Embedding === Multivariate Embedding recognizes that time-delay embeddings are not the only valid state-space construction. In Simplex and S-Map one can generate a state-space from observational vectors, or time-delay embeddings of a single observational time series, or both. === Convergent Cross Mapping === Convergent cross mapping (CCM) leverages a corollary to the Generalized Takens Theorem that it should be possible to cross predict or cross map between variables observed from the same system. Suppose that in some dynamical system involving variables X {\displaystyle X} and Y {\displaystyle Y} , X {\displaystyle X} causes Y {\displaystyle Y} . Since X {\displaystyle X} and Y {\displaystyle Y} belong to the same dynamical system, their reconstructions (via embeddings) M x {\displaystyle M_{x}} , and M y {\displaystyle M_{y}} , also map to the same system. The causal variable X {\displaystyle X} leaves a signature on the affected variable Y {\displaystyle Y} , and consequently, the reconstructed states based on Y {\displaystyle Y} can be used to cross predict values of X {\displaystyle X} . CCM leverages this property to infer causality by predicting X {\displaystyle X} using the M y {\displaystyle M_{y}} library of points (or vice versa for the other direction of causality), while assessing improvements in cross map predictability as larger and larger random samplings of M y {\displaystyle M_{y}} are used. If the prediction skill of X {\displaystyle X} increases and saturates as the entire M y {\displaystyle M_{y}} is used, this provides evidence that X {\displaystyle X} is casually influencing Y {\displaystyle Y} . === Multiview Embedding === Multiview Embedding is a Dimensionality reduction technique where a large number of state-space time series vectors are combitorially assessed towards maximal model predictability. == Extensions == Extensions to EDM techniques include: Generalized Theorems for Nonlinear State Space Reconstruction Extended Convergent Cross Mapping Dynamic stability S-Map regularization Visual analytics with EDM Convergent Cross Sorting Expert system with EDM hybrid Sliding windows based on the extended convergent cross-mapping Empirical Mode Modeling Accounting for missing data and variable step sizes Accounting for observation noise Hierarchical Bayesian EDM via Gaussian processes Intelligent and Adaptive Control Optimal control via Empirical dynamic programming Multiview distance regularised S-map

Evaluation of binary classifiers

Evaluation of a binary classifier typically assigns a numerical value, or values, to a classifier that represent its accuracy. An example is error rate, which measures how frequently the classifier makes a mistake. There are many metrics that can be used; different fields have different preferences. For example, in medicine sensitivity and specificity are often used, while in computer science precision and recall are preferred. An important distinction is between metrics that are independent of the prevalence or skew (how often each class occurs in the population), and metrics that depend on the prevalence – both types are useful, but they have very different properties. Often, evaluation is used to compare two methods of classification, so that one can be adopted and the other discarded. Such comparisons are more directly achieved by a form of evaluation that results in a single unitary metric rather than a pair of metrics. == Contingency table == Given a data set, a classification (the output of a classifier on that set) gives two numbers: the number of positives and the number of negatives, which add up to the total size of the set. To evaluate a classifier, one compares its output to another reference classification – ideally a perfect classification, but in practice the output of another gold standard test – and cross tabulates the data into a 2×2 contingency table, comparing the two classifications. One then evaluates the classifier relative to the gold standard by computing summary statistics of these 4 numbers. Generally these statistics will be scale invariant (scaling all the numbers by the same factor does not change the output), to make them independent of population size, which is achieved by using ratios of homogeneous functions, most simply homogeneous linear or homogeneous quadratic functions. Say we test some people for the presence of a disease. Some of these people have the disease, and our test correctly says they are positive. They are called true positives (TP). Some have the disease, but the test incorrectly claims they don't. They are called false negatives (FN). Some don't have the disease, and the test says they don't – true negatives (TN). Finally, there might be healthy people who have a positive test result – false positives (FP). These can be arranged into a 2×2 contingency table (confusion matrix), conventionally with the test result on the vertical axis and the actual condition on the horizontal axis. These numbers can then be totaled, yielding both a grand total and marginal totals. Totaling the entire table, the number of true positives, false negatives, true negatives, and false positives add up to 100% of the set. Totaling the columns (adding vertically) the number of true positives and false positives add up to 100% of the test positives, and likewise for negatives. Totaling the rows (adding horizontally), the number of true positives and false negatives add up to 100% of the condition positives (conversely for negatives). The basic marginal ratio statistics are obtained by dividing the 2×2=4 values in the table by the marginal totals (either rows or columns), yielding 2 auxiliary 2×2 tables, for a total of 8 ratios. These ratios come in 4 complementary pairs, each pair summing to 1, and so each of these derived 2×2 tables can be summarized as a pair of 2 numbers, together with their complements. Further statistics can be obtained by taking ratios of these ratios, ratios of ratios, or more complicated functions. The contingency table and the most common derived ratios are summarized below; see sequel for details. Note that the rows correspond to the condition actually being positive or negative (or classified as such by the gold standard), as indicated by the color-coding, and the associated statistics are prevalence-independent, while the columns correspond to the test being positive or negative, and the associated statistics are prevalence-dependent. There are analogous likelihood ratios for prediction values, but these are less commonly used, and not depicted above. == Pairs of metrics == Often accuracy is evaluated with a pair of metrics composed in a standard pattern. === Sensitivity and specificity === The fundamental prevalence-independent statistics are sensitivity and specificity. Sensitivity or True Positive Rate (TPR), also known as recall, is the proportion of people that tested positive and are positive (True Positive, TP) of all the people that actually are positive (Condition Positive, CP = TP + FN). It can be seen as the probability that the test is positive given that the patient is sick. With higher sensitivity, fewer actual cases of disease go undetected (or, in the case of the factory quality control, fewer faulty products go to the market). Specificity (SPC) or True Negative Rate (TNR) is the proportion of people that tested negative and are negative (True Negative, TN) of all the people that actually are negative (Condition Negative, CN = TN + FP). As with sensitivity, it can be looked at as the probability that the test result is negative given that the patient is not sick. With higher specificity, fewer healthy people are labeled as sick (or, in the factory case, fewer good products are discarded). The relationship between sensitivity and specificity, as well as the performance of the classifier, can be visualized and studied using the Receiver Operating Characteristic (ROC) curve. In theory, sensitivity and specificity are independent in the sense that it is possible to achieve 100% in both (such as in the red/blue ball example given above). In more practical, less contrived instances, however, there is usually a trade-off, such that they are inversely proportional to one another to some extent. This is because we rarely measure the actual thing we would like to classify; rather, we generally measure an indicator of the thing we would like to classify, referred to as a surrogate marker. The reason why 100% is achievable in the ball example is because redness and blueness is determined by directly detecting redness and blueness. However, indicators are sometimes compromised, such as when non-indicators mimic indicators or when indicators are time-dependent, only becoming evident after a certain lag time. The following example of a pregnancy test will make use of such an indicator. Modern pregnancy tests do not use the pregnancy itself to determine pregnancy status; rather, human chorionic gonadotropin is used, or hCG, present in the urine of gravid females, as a surrogate marker to indicate that a woman is pregnant. Because hCG can also be produced by a tumor, the specificity of modern pregnancy tests cannot be 100% (because false positives are possible). Also, because hCG is present in the urine in such small concentrations after fertilization and early embryogenesis, the sensitivity of modern pregnancy tests cannot be 100% (because false negatives are possible). === Positive and negative predictive values === In addition to sensitivity and specificity, the performance of a binary classification test can be measured with positive predictive value (PPV), also known as precision, and negative predictive value (NPV). The positive prediction value answers the question "If the test result is positive, how well does that predict an actual presence of disease?". It is calculated as TP/(TP + FP); that is, it is the proportion of true positives out of all positive results. The negative prediction value is the same, but for negatives, naturally. ==== Impact of prevalence on predictive values ==== Prevalence has a significant impact on prediction values. As an example, suppose there is a test for a disease with 99% sensitivity and 99% specificity. If 2000 people are tested and the prevalence (in the sample) is 50%, 1000 of them are sick and 1000 of them are healthy. Thus about 990 true positives and 990 true negatives are likely, with 10 false positives and 10 false negatives. The positive and negative prediction values would be 99%, so there can be high confidence in the result. However, if the prevalence is only 5%, so of the 2000 people only 100 are really sick, then the prediction values change significantly. The likely result is 99 true positives, 1 false negative, 1881 true negatives and 19 false positives. Of the 19+99 people tested positive, only 99 really have the disease – that means, intuitively, that given that a patient's test result is positive, there is only 84% chance that they really have the disease. On the other hand, given that the patient's test result is negative, there is only 1 chance in 1882, or 0.05% probability, that the patient has the disease despite the test result. === Precision and recall === Precision and recall can be interpreted as (estimated) conditional probabilities: Precision is given by P ( C = P | C ^ = P ) {\displaystyle P(C=P|{\hat {C}}=P)} while recall is given by P ( C ^ = P | C = P ) {\displaystyle P({\hat {C}}=P|C=P)} , where C ^ {\

Language model benchmark

A language model benchmark is a standardized test designed to evaluate the performance of language models on various natural language processing tasks. These tests are intended for comparing different models' capabilities in areas such as language understanding, generation, and reasoning. Benchmarks generally consist of a dataset and corresponding evaluation metrics. The dataset provides text samples and annotations, while the metrics measure a model's performance on tasks like answering questions, text classification, and machine translation. These benchmarks are developed and maintained by academic institutions, research organizations, and industry players to track progress in the field. In addition to accuracy, the metrics can include throughput, energy efficiency, bias, trust, and sustainability. == Overview == === Types === Benchmarks may be described by the following adjectives, not mutually exclusive: Classical: These tasks are studied in natural language processing, even before the advent of deep learning. Examples include the Penn Treebank for testing syntactic and semantic parsing, as well as bilingual translation benchmarked by BLEU scores. Question answering: These tasks have a text question and a text answer, often multiple-choice. They can be open-book or closed-book. Open-book QA resembles reading comprehension questions, with relevant passages included as annotation in the question, in which the answer appears. Closed-book QA includes no relevant passages. Closed-book QA is also called open-domain question-answering. Before the era of large language models, open-book QA was more common, and understood as testing information retrieval methods. Closed-book QA became common since GPT-2 as a method to measure knowledge stored within model parameters. Omnibus: An omnibus benchmark combines many benchmarks, often previously published. It is intended as an all-in-one benchmarking solution. Reasoning: These tasks are usually in the question-answering format, but are intended to be more difficult than standard question answering. Multimodal: These tasks require processing not only text, but also other modalities, such as images and sound. Examples include OCR and transcription. Agency: These tasks are for a language-model–based software agent that operates a computer for a user, such as editing images, browsing the web, etc. Adversarial: A benchmark is "adversarial" if the items in the benchmark are picked specifically so that certain models do badly on them. Adversarial benchmarks are often constructed after state of the art (SOTA) models have saturated (achieved 100% performance) a benchmark, to renew the benchmark. A benchmark is "adversarial" only at a certain moment in time, since what is adversarial may cease to be adversarial as newer SOTA models appear. Public/Private: A benchmark might be partly or entirely private, meaning that some or all of the questions are not publicly available. The idea is that if a question is publicly available, then it might be used for training, which would be "training on the test set" and invalidate the result of the benchmark. Usually, only the guardians of the benchmark have access to the private subsets, and to score a model on such a benchmark, one must send the model weights, or provide API access, to the guardians. The boundary between a benchmark and a dataset is not sharp. Generally, a dataset contains three "splits": training, test, and validation. Both the test and validation splits are essentially benchmarks. In general, a benchmark is distinguished from a test/validation dataset in that a benchmark is typically intended to be used to measure the performance of many different models that are not trained specifically for doing well on the benchmark, while a test/validation set is intended to be used to measure the performance of models trained specifically on the corresponding training set. In other words, a benchmark may be thought of as a test/validation set without a corresponding training set. Conversely, certain benchmarks may be used as a training set, such as the English Gigaword or the One Billion Word Benchmark, which in modern language is just the negative log-likelihood loss on a pretraining set with 1 billion words. Indeed, the distinction between benchmark and dataset in language models became sharper after the rise of the pretraining paradigm, whereby a model is first trained on massive, unlabeled datasets to learn general language patterns, syntax, and knowledge (pretraining), and the base model is then adapted to specific, downstream tasks using smaller, labeled datasets (fine-tuning). === Lifecycle === Generally, the life cycle of a benchmark consists of the following steps: Inception: A benchmark is published. It can be simply given as a demonstration of the power of a new model (implicitly) that others then picked up as a benchmark, or as a benchmark that others are encouraged to use (explicitly). Growth: More papers and models use the benchmark, and the performance on the benchmark grows. Maturity, degeneration or deprecation: A benchmark may be saturated, after which researchers move on to other benchmarks. Progress on the benchmark may also be neglected as the field moves to focus on other benchmarks. Renewal: A saturated benchmark can be upgraded to make it no longer saturated, allowing further progress. === Construction === Like datasets, benchmarks are typically constructed by several methods, individually or in combination: Web scraping: Ready-made question-answer pairs may be scraped online, such as from websites that teach mathematics and programming. Conversion: Items may be constructed programmatically from scraped web content, such as by blanking out named entities from sentences, and asking the model to fill in the blank. This was used for making the CNN/Daily Mail Reading Comprehension Task. Crowd sourcing: Items may be constructed by paying people to write them, such as on Amazon Mechanical Turk. This was used for making the MCTest. === Evaluation === Generally, benchmarks are fully automated. This limits the questions that can be asked. For example, with mathematical questions, "proving a claim" would be difficult to automatically check, while "calculate an answer with a unique integer answer" would be automatically checkable. With programming tasks, the answer can generally be checked by running unit tests, with an upper limit on runtime. The benchmark scores are of the following kinds: For multiple choice or cloze questions, common scores are accuracy (frequency of correct answer), precision, recall, F1 score, etc. pass@n: The model is given n {\displaystyle n} attempts to solve each problem. If any attempt is correct, the model earns a point. The pass@n score is the model's average score over all problems. k@n: The model makes n {\displaystyle n} attempts to solve each problem, but only k {\displaystyle k} attempts out of them are selected for submission. If any submission is correct, the model earns a point. The k@n score is the model's average score over all problems. cons@n: The model is given n {\displaystyle n} attempts to solve each problem. If the most common answer is correct, the model earns a point. The cons@n score is the model's average score over all problems. Here "cons" stands for "consensus" or "majority voting". The pass@n score can be estimated more accurately by making N > n {\displaystyle N>n} attempts, and use the unbiased estimator 1 − ( N − c n ) ( N n ) {\displaystyle 1-{\frac {\binom {N-c}{n}}{\binom {N}{n}}}} , where c {\displaystyle c} is the number of correct attempts. For less well-formed tasks, where the output can be any sentence, there are the following commonly used scores including BLEU ROUGE, METEOR, NIST, word error rate, LEPOR, CIDEr, and SPICE. === Issues === error: Some benchmark answers may be wrong. ambiguity: Some benchmark questions may be ambiguously worded. subjective: Some benchmark questions may not have an objective answer at all. This problem generally prevents creative writing benchmarks. Similarly, this prevents benchmarking writing proofs in natural language, though benchmarking proofs in a formal language is possible. open-ended: Some benchmark questions may not have a single answer of a fixed size. This problem generally prevents programming benchmarks from using more natural tasks such as "write a program for X", and instead uses tasks such as "write a function that implements specification X". inter-annotator agreement: Some benchmark questions may be not fully objective, such that even people would not agree with 100% on what the answer should be. This is common in natural language processing tasks, such as syntactic annotation. shortcut: Some benchmark questions may be easily solved by an "unintended" shortcut. For example, in the SNLI benchmark, having a negative word like "not" in the second sentence is a strong signal for the "Contradiction" category, regardless of what the se

List of publications in data science

Timeline of artificial intelligence risks in global finance

Empirical dynamic modeling

Evaluation of binary classifiers

Language model benchmark

Empirical dynamic modeling

Related Articles