Paper debiases multiple word embedding biases simultaneously.
problem Reduction of multiple biases in word embeddings.
method Joint multiclass debiasing approach using Word Embeddings Association Test (WEAT).
result Demonstrates reduction or complete elimination of bias in word embeddings.
This paper evaluates debiasing methods on word embeddings to reduce religious bias.
problem Social biases persist in word embeddings, potentially amplifying them in AI applications.
method Investigates and evaluates three multiclass debiasing techniques on three word embeddings.
result ConceptorDebiasing is the most effective method, reducing religious bias by 82-96%.
New method neutralizes gender bias in word embeddings without losing semantic information.
problem Gender biases in word embeddings trained on human-generated corpora.
method Latent Disentanglement and Counterfactual Generation with siamese auto-encoder and gradient reversal layer.
result Our method outperforms existing debiasing methods in preserving semantic information and neutralizing gender biases.
This paper debiases KG embeddings by filtering sensitive attributes.
problem Social and cultural biases in KG representations.
method Exploring and filtering popularity and gender biases in KGE.
result Sensitivity of KG embeddings to sensitive attributes like gender.
MONET debiases graph embeddings by training on metadata-orthogonal dimensions.
problem Graph embeddings can be biased by node attributes, affecting fairness and interpretability.
method MONET trains embeddings on a hyperplane orthogonal to node metadata.
result MONET effectively removes bias from node embeddings, improving fairness and interpretability.
DeBayes uses Bayesian methods to create fair network embeddings.
problem Ensuring fairness in network embeddings for high-impact applications.
method Bayesian approach to learn debiased network embeddings.
result DeBayes produces fairer network embeddings for link prediction.
Paper debiases text embeddings using context injection.
problem Text embeddings capture and perpetuate biases.
method Review 19 embedding models, quantify biases, and test context injection.
result Higher performing models are more prone to biases but better at incorporating context.
(Bolukbasi et al., 2016) demonstrated that pretrained word embeddings can inherit gender bias from the data they were trained on. We investigate how this bias affects downstream classification tasks, using the case study of occupation classification (De-Arteaga et al.,2019). We show that traditional techniques for debi…
Online texts -- across genres, registers, domains, and styles -- are riddled with human stereotypes, expressed in overt or subtle ways. Word embeddings, trained on these texts, perpetuate and amplify these stereotypes, and propagate biases to machine learning models that use word embeddings as features. In this work, w…
Unified theory and debiasing framework for random oblique projections in high dimensions.
problem Systematic statistical bias in random oblique projections induced by sampling.
method Unified non-asymptotic theory and debiasing framework.
result Sharp bias--variance characterizations and improved approximation accuracy.
Algorithm D4 decomposes data into useful and orthogonal components.
problem Data representation and task-specific information.
method Decision-Directed Data Decomposition (D4) algorithm.
result Improves predictive generalization and debiasing in word embeddings.
Proposes a three-stage debiasing framework to improve out-of-distribution accuracy.
problem Inaccurate uncertainty estimations in bias-only models damage ensemble-based debiasing performance.
method Calibrates the bias-only model to improve its uncertainty estimations, creating a three-stage ensemble-based debiasing framework.
result The three-stage debiasing framework consistently outperforms traditional methods in out-of-distribution accuracy.
Paper develops PGMM framework for debiased inference on nonparametric IV estimators.
problem Automatic debiased inference on nonparametric IV functionals.
method Penalized GMM (PGMM) framework for functionals of IV estimators.
result PGMM-based debiased estimator performs well, achieving near-nominal coverage.
D3M debiases models by selectively removing problematic examples.
problem Model failures on underrepresented subgroups.
method Isolates and removes specific training examples that cause failures.
result Efficiently trains debiased classifiers with minimal example removal.
A new method debiases multiple target parameters without IFs.
problem Debiasing multiple target parameters in nonparametric models.
method Kernel Debiased Plug-in Estimation (KDPE) using TMLE and reproducing kernel Hilbert spaces.
result KDPE simultaneously debiases all pathwise differentiable target parameters.
A self-supervised debiasing method using rank regularization mitigates spurious correlations in neural networks.
problem Spurious correlations cause biases in deep neural networks, affecting generalization.
method Spectral analysis of latent representations, rank regularization, self-supervised pretraining, debiasing of downstream tasks.
result The proposed framework significantly improves generalization performance and outperforms supervised debiasing approaches.
A new debiasing method for high-dimensional regression with applications to PCR.
problem Debiasing in high-dimensional statistics with i.i.d. samples and sub-Gaussian covariates.
method Spectrum-Aware Debiasing using rescaled gradient descent with spectral information.
result Achieves debiasing in broader contexts with structured dependencies, heavy tails, and low-rank structures.
Proposes online debiasing to correct bias in adaptive data collection for high-dimensional linear regression.
problem Bias in adaptive data collection for high-dimensional linear regression.
method Online debiasing procedure for LASSO and other estimators.
result Optimal debiasing of LASSO estimator in specific sparsity regime.
A debiasing method improves nonparametric regression's statistical properties.
problem Lack of theoretical guarantees for modern nonparametric regression methods.
method Model-free debiasing method incorporating a correction term.
result Debiased estimator satisfies pointwise and uniform risk convergence, asymptotic normality.
Study improves statistical inference for CATEs using Lasso and DML.
problem Estimating and inferring CATEs in high-dimensional settings.
method Doubly robust estimator, Lasso regularization, debiased Lasso, DML.
result TDL (triple/debiased Lasso) achieves n \sqrt{n} n -consistency and confidence intervals. Develops a direct debiased machine learning framework using Bregman divergence.
problem Reduces bias in machine learning estimates of causal effects or structural models.
method Neyman targeted estimation and generalized Riesz regression using Bregman divergence.
result Improves estimation of parameters of interest in causal models.
Ridge regression is revisited with debiasing and thresholding, offering advantages over Lasso.
problem High-dimensional data challenges classical ridge regression's sparsity detection and bias issues.
method Debiasing and thresholding ridge regression, introducing a wild bootstrap for confidence regions and hypothesis testing, and a hybrid bootstrap for prediction intervals.
result Debiased and thresholded ridge regression can offer similar performance to thresholded Lasso and may be preferable in some settings.
DKGM improves image quality by debiasing kernel-based models.
problem Improving image quality in generative models without losing detail.
method Two-stage framework: KDE for density estimation, debiasing algorithm inspired by stochastic approximation.
result DKGM achieves comparable image quality to state-of-the-art models on CIFAR10.
Study examines biases in clinical word embeddings, revealing performance gaps across groups.
problem Biases in clinical word embeddings leading to performance differences across groups.
method Pretrained BERT models on MIMIC-III, fill-in-the-blank method, fairness evaluation on clinical tasks.
result Classifiers trained from BERT representations exhibit statistically significant differences in performance across groups.
Python package automates causal parameter estimation using Riesz regression.
problem Efficient estimation of causal and structural parameters.
method Automatic DML and generalized Riesz regression framework.
result Automatic construction of balancing link functions for generalized Riesz regression.
Paper develops a distributed debiased estimator for sparse statistical inference.
problem High computational costs in debiased estimator construction for high-dimensional models.
method Develops a multi-round distributed debiased estimator using both labeled and unlabelled data.
result Unlabeled data improves statistical rate of each iteration in distributed setup.
We simplify debiasing of LASSO estimates for practical computation.
problem Bias in LASSO estimates for high-dimensional data.
method Reformulate optimization problem to compute debiasing matrix W directly.
result Efficient closed-form solution for W with unique optimal solution.
COMMOD debiases models with minimal and interpretable changes.
problem Inconsistent and costly model updates in fair machine learning.
method Introduced COMMOD, a novel algorithm for algorithmic fairness that minimizes changes and makes them interpretable.
result COMMOD achieves comparable performance to state-of-the-art debiasing methods while making minimal and interpretable changes.
Bayesian inference corrected for bias in high-dimensional models.
problem Bayesian inference for high-dimensional regression models often produces biased credible sets.
method Debiasing approach based on Bernstein-von Mises theorem.
result Frequentist validity of debiased Bayesian posterior.
A theorem for debiasing machine learning with finite sample guarantees.
problem Calculating confidence intervals for machine learning functionals.
method Debiased machine learning based on bias correction and sample splitting.
result Nonasymptotic debiased machine learning theorem with finite sample guarantees.
New method improves robustness of neural network-based debiasing.
problem Improving robustness of neural network-based debiasing.
method Moment-constrained learning for neural networks.
result Improved performance compared to state-of-the-art benchmarks.
ADIGen: Automatic, Debiased, and Invariant Counterfactual Generation
problem Generative models for counterfactual outcomes
method ADIGen combines Riesz regression, causal invariance, and orthogonal statistical learning
result ADIGen controls counterfactual risk under general interventions
Mitigates confirmation bias in SSL by adjusting pseudo labels dynamically.
problem Confirmation bias in semi-supervised learning leads to errors in pseudo labels.
method TaMatch framework adjusts scaling ratio to debias pseudo labels and dynamically adjusts target distribution.
result TaMatch significantly outperforms existing methods in SSL tasks.
Automatic debiasing for causal and policy effects using Neural Nets and Random Forests.
problem Estimating causal and policy effects from high-dimensional or non-parametric regression functions.
method Automatic learning of Riesz representation using Neural Nets and Random Forests.
result Automatic debiasing method performs well compared to state-of-the-art algorithms.
New debiasing methods for fine-tuned neural networks.
problem Bias in neural networks used for high-stakes decisions.
method Intra-processing methods designed for fine-tuned models.
result Demonstrated effectiveness of intra-processing methods across various datasets.
MAPPING debiases GNNs for fair node classification with limited leakage.
problem Graph Neural Networks inherit and exacerbate historical discrimination in high-stake domains.
method MAPPING uses distance covariance-based fairness constraints and adversarial debiasing.
result MAPPING achieves better trade-offs between fairness and utility, mitigating privacy risks.
Javanmard and Montanari propose a debiased estimator for high-dimensional regression.
problem Bias in high-dimensional regression models.
method Debiased LASSO estimator.
result Debiased LASSO yields asymptotically normal estimators and valid hypothesis tests.
New method for debiased inference without assuming exact solutions in inverse problems.
problem Dealing with inverse problems where exact solutions may not exist.
method Nonparametric instrumental variable analysis without structural equations.
result Valid inference on functionals of inverse problems without assuming exact solutions.
The blind application of machine learning runs the risk of amplifying biases present in data. Such a danger is facing us with word embedding, a popular framework to represent text data as vectors which has been used in many machine learning and natural language processing tasks. We show that even word embeddings traine…
New models can't beat existing ones, so debiasing methods only slightly reduce needed labels.
problem Limiting scalability in model evaluation due to self-preferencing biases.
method Study of debiasing methods using a few high-quality labels to reduce model judgments.
result Debiasing methods can't decrease required ground truth labels by more than half when the judge is no more accurate than the model.
ScoreMatchingRiesz improves debiased machine learning and policy effects estimation.
problem Improving debiased machine learning and policy effects estimation.
method Score matching and Riesz representer estimation.
result Estimates policy path for continuous treatments, improving interpretability.
Study evaluates gender bias in relation extraction systems.
problem Gender bias in relation extraction systems.
method Created WikiGenderBias dataset, evaluated systems for bias, analyzed bias mitigation techniques.
result NRE systems exhibit gender bias in predictions.
New method debiases selection bias in PU classification with exposure data.
problem Binary classification from positive and unlabeled data with selection bias.
method Automatic Debiased PUE (ADPUE) learning method.
result ADPUE outperforms traditional PU learning methods on various datasets.
Semi-supervised learning debiased for better performance.
problem Bias in SSL methods impairs theoretical guarantees and empirical performance.
method Simple debiasing approach to remove risk estimate bias.
result Debiased SSL methods provide better calibrated models.
New MLMC method reduces evidence estimation cost.
problem Efficiently estimating model evidence in Bayesian inference.
method Multilevel Monte Carlo (MLMC) sampling for unbiased estimation.
result Significant computational savings in estimating model evidence.
The paper tackles biases in session-based recommender systems by modeling user interest as a stochastic process.
problem Data uncertainty, popularity bias, and exposure bias in session-based recommender systems.
method The paper proposes treating user interest as a stochastic process in the latent space, debiasing item embeddings, modeling dense user interest, and introducing fake targets to simulate extended exposure.
result The proposed approach mitigates challenges in session-based recommender systems, as shown by computational experiments on various datasets.
Proposes debiasing strategy for ill-posed regression problems.
problem Estimating functions with conditional moment restrictions, especially when estimators are sensitive to misspecification.
method Debiased estimation using influence function of modified mean squared error.
result Demonstrates finite-sample convergence rate and robustness to misspecification.
Improved nonparametric regression with debiasing for root-n consistency.
problem Challenges in achieving root-n consistency and normal distribution for nonparametric estimators.
method Debiasing technique by adding a correction term to nonparametric estimators.
result Achieves root-n consistency and asymptotic normality.