Study improves statistical power for detecting algorithmic bias in educational data.
problem Challenges in measuring algorithmic bias using ABROCA due to skewed distribution.
method Investigates ABROCA's distributional properties and proposes nonparametric randomization tests.
result ABROCA-based bias assessments are underpowered in typical EDM sample sizes.
Machine learning forecasts show bias at long horizons, contrary to standard tests.
problem Forecast efficiency tests misinterpret machine learning performance.
method Theoretical and empirical analysis of regularization and measurement noise.
result Machine learning forecasts exhibit overreaction at longer horizons, not bias.
Unified framework for mean testing under truncation bias.
problem High-dimensional mean testing under arbitrary truncation.
method Characterizes fundamental limits and develops a simple second-order test.
result Unified framework connects finite-moment, sub-Gaussian, and median-regular structural regimes.
Adversarial training leads to large generalization gap, decomposed into bias and variance.
problem Understanding the large generalization gap in adversarially trained models.
method Bias-Variance decomposition of test risk as a function of adversarial perturbation radius.
result Bias increases monotonically with adversarial perturbation radius and is dominant in test risk.
Study evaluates bias mitigation methods in deep learning, finds they often exploit hidden biases.
problem Deep learning systems learn biases, affecting performance on minority groups.
method Improved evaluation protocol, new dataset, robustness across different tuning distributions.
result Bias mitigation methods often exploit hidden biases, are not robust to multiple forms of bias, and are sensitive to tuning set choice.
Proposes tests to control confounding bias in predictive models.
problem Lack of non-parametric tests for confounding bias in predictive modeling.
method Partial and full confounder tests for probing null hypotheses of unconfounded and fully confounded models.
result Reveals previously unreported or hard-to-correct confounders in machine learning models.
Estimates peeking effects in p-values to correct bias.
problem Data peeking biases reported p-values downward.
method Develops mechanisms to estimate running extrema of test statistics.
result Corrects bias in p-values due to peeking.
The bias-variance tradeoff tells us that as model complexity increases, bias falls and variances increases, leading to a U-shaped test error curve. However, recent empirical results with over-parameterized neural networks are marked by a striking absence of the classic U-shaped test error curve: test error keeps decrea…
New framework minimizes interference and selection bias in network A/B testing.
problem Interference and selection bias in network A/B testing.
method Proposes a principled framework that jointly minimizes interference and selection bias using edge spillover probability and cluster matching.
result Significantly lower error in causal effect estimation compared to existing solutions.
Active testing reduces label costs for efficient model evaluation.
problem Real-world applications require expensive test labels, disconnecting from existing model evaluation methods.
method Derives acquisition strategies to select test points efficiently, addressing label bias and variance.
result Active testing improves model evaluation efficiency without sacrificing accuracy.
In real-world classification problems, the class balance in the training dataset does not necessarily reflect that of the test dataset, which can cause significant estimation bias. If the class ratio of the test dataset is known, instance re-weighting or resampling allows systematical bias correction. However, learning…
The paper develops a theory explaining how machine learning models can amplify biases.
problem Understanding and mitigating bias in machine learning models.
method Analytical theory of ridge regression with and without random projections.
result Observations and predictions align with empirical data on machine learning bias.
Study shows different trajectory prediction models generalize better under OoD conditions.
problem Comparing trajectory prediction models' robustness across different datasets.
method Training models on Argoverse 2 and testing on Waymo Open Motion, and vice versa, with various augmentation strategies.
result Smallest model with highest inductive bias performs best in OoD generalization.
New insights on bias in multi-armed bandits under conditional sampling.
problem Understanding bias in multi-armed bandits under conditional sampling.
method Characterized the sign of conditional bias of monotone functions of rewards.
result Sign of conditional bias can differ from marginal bias, depending on conditioning events.
In classification problems, sampling bias between training data and testing data is critical to the ranking performance of classification scores. Such bias can be both unintentionally introduced by data collection and intentionally introduced by the algorithm, such as under-sampling or weighting techniques applied to i…
Deep neural networks can generalize by reducing high-frequency noise over time, not always following a monotonic learning bias.
problem Understanding the learning dynamics and generalization of over-parameterized DNNs.
method Experimental analysis of deep double descent, focusing on the spectral bias of DNNs.
result The high-frequency components of DNNs diminish over training, leading to a second descent in test error.
Propensity score matching improves fairness in machine learning models.
problem Bias in training data affects fairness metrics in machine learning models.
method Propensity score matching to evaluate and mitigate bias in test data.
result FairMatch significantly reduces bias in test data without sacrificing predictive performance.
New tools for assessing and correcting bias in AI algorithms.
problem Fairness and bias in AI algorithms, especially when ground truth data is unavailable.
method Three tools: controlled fairness, retraining algorithms, and parameter adjustment algorithms.
result Effective in reducing bias and improving fairness in AI models.
Synthetic control method improves policy evaluation in high-dimensional settings.
problem Evaluating the impact of new policies in large-scale applications.
method Two-phase approach: nearest neighbor matching followed by supervised learning.
result The method successfully improves estimate accuracy in large-scale experiments.
For many causal effect parameters of interest, doubly robust machine learning (DRML) estimators ψ^1 are the state-of-the-art, incorporating the good prediction performance of machine learning; the decreased bias of doubly robust estimators; and the analytic tractability and bias reduction of sample splitting wi…
Performance of investment managers are evaluated in comparison with benchmarks, such as financial indices. Due to the operational constraint that most professional databases do not track the change of constitution of benchmark portfolios, standard tests of performance suffer from the "look-ahead benchmark bias," when t…
Framework tests group fairness in machine learning models.
problem Detecting biases in machine learning classifiers.
method Optimal transport projections to audit group fairness.
result Statistical test for various fairness notions efficiently computed.
With recent advances in high throughput technology, researchers often find themselves running a large number of hypothesis tests (thousands+) and esti- mating a large number of effect-sizes. Generally there is particular interest in those effects estimated to be most extreme. Unfortunately naive estimates of these effe…
Deep networks generalize well even when they fit training data perfectly, thanks to overparametrization.
problem Understanding generalization in overparametrized deep networks.
method Random features regression, asymptotic analysis, ensemble averaging.
result Bias remains constant beyond the interpolation threshold, while variance components decay with overparametrization.
This paper examines the use of a residual bootstrap for bias correction in machine learning regression methods. Accounting for bias is an important obstacle in recent efforts to develop statistical inference for machine learning methods. We demonstrate empirically that the proposed bootstrap bias correction can lead to…
We detect lookahead bias in LLM forecasts using a novel statistical method.
problem Detecting lookahead bias in LLM-generated economic forecasts.
method Developed a statistical procedure using date-only recall queries and estimated Lookahead Propensity (LAP).
result LLM forecasts are contaminated with lookahead bias, as indicated by a positive interaction between LAP and the forecast in accuracy regressions.
New insights into bias and variance in over-parameterized models.
problem Understanding bias and variance in over-parameterized models.
method Analytic expressions derived from statistical physics for two minimal models.
result Over-parameterized models can overfit even in noiseless conditions.
Securely trains fair models using homomorphic encryption.
problem Protecting sensitive features while testing model fairness.
method Fully homomorphic encryption for training and testing.
result Practical application to adult income data set.
Corrects bias in LLM-as-a-judge evaluations using adaptive calibration.
problem Bias in LLM evaluations due to imperfect sensitivity and specificity.
method Plug-in framework with confidence intervals accounting for test and calibration dataset uncertainties.
result LML-based evaluation yields more reliable estimates than human-only evaluation.
Develops tools to audit ML models for bias and unfairness.
problem Auditing ML models for individual bias and unfairness.
method Formalizes the task as an optimization problem and develops inferential tools for the optimal value.
result Demonstrates the utility of tools in revealing biases in COMPAS recidivism prediction instrument.
The paper detects and identifies bias in data using a counterfactual approach.
problem Detecting and identifying bias in data, especially in medical image classification.
method A global explanation framework using the counterfactual approach to identify bias causing artifacts.
result Black frames significantly influence Convolutional Neural Network's prediction, changing benign to malignant.
Proposes a neural network model to improve predictions in biased datasets.
problem Reduces bias in predictions from datasets with selection bias.
method Integrates meta information about bias direction and quantity into a neural network model.
result Improves prediction accuracy in electoral polls with biased training data.
Most of previous machine learning algorithms are proposed based on the i.i.d. hypothesis. However, this ideal assumption is often violated in real applications, where selection bias may arise between training and testing process. Moreover, in many scenarios, the testing data is not even available during the training pr…
Recent developments in Neural Relation Extraction (NRE) have made significant strides towards Automated Knowledge Base Construction (AKBC). While much attention has been dedicated towards improvements in accuracy, there have been no attempts in the literature to our knowledge to evaluate social biases in NRE systems. W…
The accuracy of deep neural networks is significantly affected by how well mini-batches are constructed during the training step. In this paper, we propose a novel adaptive batch selection algorithm called Recency Bias that exploits the uncertain samples predicted inconsistently in recent iterations. The historical lab…
The paper introduces Relative Bias to quantify LLM bias systematically.
problem Quantifying bias in LLMs is challenging due to ambiguity and rapid model emergence.
method Relative Bias framework using Embedding Transformation and LLM-as-a-Judge methodologies.
result The two scoring methods show strong alignment, providing a systematic approach.
DCEM algorithm reduces bias in machine learning models trained on selective labels.
problem Bias in machine learning models trained on selective labels.
method Disparate Censorship Expectation-Maximization (DCEM) algorithm.
result DCEM improves bias mitigation without sacrificing discriminative performance.
Unintended bias in Machine Learning can manifest as systemic differences in performance for different demographic groups, potentially compounding existing challenges to fairness in society at large. In this paper, we introduce a suite of threshold-agnostic metrics that provide a nuanced view of this unintended bias, by…
Paper addresses selection bias in online advertising auctions.
problem Selection bias affects auction truthfulness and advertiser profits.
method Theoretical analysis combined with multi-task learning.
result Selection bias can be significantly reduced using multi-task learning.
Excessive reuse of holdout data can lead to overfitting. However, there is little concrete evidence of significant overfitting due to holdout reuse in popular multiclass benchmarks today. Known results show that, in the worst-case, revealing the accuracy of k adaptively chosen classifiers on a data set of size n al…
Data that is gathered adaptively --- via bandit algorithms, for example --- exhibits bias. This is true both when gathering simple numeric valued data --- the empirical means kept track of by stochastic bandit algorithms are biased downwards --- and when gathering more complicated data --- running hypothesis tests on c…
Standard methods in supervised learning separate training and prediction: the model is fit independently of any test points it may encounter. However, can knowledge of the next test point x⋆ be exploited to improve prediction accuracy? We address this question in the context of linear prediction, show…
From scientific experiments to online A/B testing, the previously observed data often affects how future experiments are performed, which in turn affects which data will be collected. Such adaptivity introduces complex correlations between the data and the collection procedure. In this paper, we prove that when the dat…
Bayesian method corrects bias in imbalanced datasets.
problem Prevalence bias in machine learning datasets.
method Bayesian risk minimization framework, bias-corrected loss function.
result Corrected loss function improves model performance.
Social bias in machine learning has drawn significant attention, with work ranging from demonstrations of bias in a multitude of applications, curating definitions of fairness for different contexts, to developing algorithms to mitigate bias. In natural language processing, gender bias has been shown to exist in contex…
Publication bias skews asset pricing research findings.
problem Bias in sharing and publishing research findings.
method Meta-studies and empirical Bayes corrections.
result Publication bias effects are minimal and not dominant.
Proposes BSSP to stabilize predictions in biased data.
problem Distribution shift between training and test data causes prediction instability.
method Balance-subsampled stable prediction (BSSP) algorithm based on fractional factorial design.
result Significantly improves prediction stability across unknown test data.
Algorithm recovers causal graphs in presence of latent confounders and selection bias.
problem Recovering causal graphs in the presence of latent confounders and selection bias.
method Iterative causal discovery (ICD) algorithm that relies on causal Markov and faithfulness assumptions.
result Sound and complete algorithm that recovers the equivalence class of the underlying causal graph.