Estimation Bias

Estimation bias, the systematic difference between an estimated value and its true value, is a pervasive problem across diverse fields of machine learning and statistics. Current research focuses on developing methods to mitigate this bias in various contexts, including reinforcement learning (e.g., through techniques like twin delayed deep deterministic policy gradient and adaptive ensemble Q-learning), causal inference (using inverse probability weighting and counterfactual approaches), and high-dimensional data analysis (employing techniques like moment-constrained learning and Pareto-smoothed weighting). Addressing estimation bias is crucial for improving the reliability and accuracy of models and algorithms across numerous applications, ranging from medical diagnosis to economic forecasting and autonomous systems.

Papers

September 29, 2024

Automatic debiasing of neural networks via moment-constrained learning
Christian L. Hines, Oliver J. Hines
Neural Network Conditional Density Estimation Bias Data Debiasing Effect Estimation

May 24, 2024

Information-theoretic Generalization Analysis for Expected Calibration Error
Futoshi Futami, Masahiro Fujisawa
Calibration Performance Generalization Error Bias Evaluation Calibration Error Estimation Bias Information Theoretic Generalization

May 23, 2024

Exclusively Penalized Q-learning for Offline Reinforcement Learning
Junghyuk Yeom, Yonghyeon Jo, Jungmo Kim, Sanghyeon Lee, Seungyul Han
Offline Reinforcement Learning Q Learning Estimation Bias Policy Constraint Regularized Reinforcement Learning

April 26, 2024

February 14, 2024

Exploiting Estimation Bias in Deep Double Q-Learning for Actor-Critic Methods
Alberto Sinigaglia, Niccolò Turcato, Alberto Dalla Libera, Ruggero Carli, Gian Antonio Susto
Reinforcement Learning Policy Learning Actor Critic Estimation Bias Double Deep Q Learning Delayed Deep Deterministic Policy Gradient

January 4, 2024

Disentangle Estimation of Causal Effects from Cross-Silo Data
Yuxuan Liu, Haozhao Wang, Shuang Wang, Zhiming He, Wenchao Xu, Jialiang Zhu, Fan Yang
Causal Effect Disentanglement Capability Causal Effect Estimation Estimation Bias Local Effect

December 24, 2023

Deep Copula-Based Survival Analysis for Dependent Censoring with Identifiability Guarantees
Weijia Zhang, Chun Kai Ling, Xuanhui Zhang
Ground Truth Estimation Bias Partial Identifiability Survival Analysis Censored Time to Event

November 7, 2023

Mitigating Estimation Errors by Twin TD-Regularized Actor and Critic for Deep Reinforcement Learning
Junmin Zhong, Ruofan Wu, Jennie Si
Deep Reinforcement Learning Distributional Learning Multiple Critic Estimation Bias Identical Twin Actor Loss Estimation Error

October 12, 2023

Debias the Training of Diffusion Models
Hu Yu, Li Shen, Jie Huang, Man Zhou, Hongsheng Li, Feng Zhao
Diffusion Model Training Data Gaussian Noise Self Debiasing Estimation Bias Bias Challenge Denoising Score Matching

September 20, 2023

RHALE: Robust and Heterogeneity-aware Accumulated Local Effects
Vasilis Gkolemis, Theodore Dalamagas, Eirini Ntoutsi, Christos Diou
Estimation Bias Heterogeneous Effect Average Causal Effect Local Effect

August 23, 2023

Bias-Aware Minimisation: Understanding and Mitigating Estimator Bias in Private SGD
Moritz Knolle, Robert Dorfman, Alexander Ziller, Daniel Rueckert, Georgios Kaissis
Human Understanding Bias Mitigation Private Gradient Estimation Bias Gradient Oracle

June 20, 2023

Adaptive Ensemble Q-learning: Minimizing Estimation Bias via Error Feedback
Hang Wang, Sen Lin, Junshan Zhang
Q Learning Error Feedback Estimation Bias Adaptive Ensemble Function Estimation Ensemble Q Learning

May 25, 2023

A Guide Through the Zoo of Biased SGD
Yury Demidovich, Grigory Malinovsky, Igor Sokolov, Peter Richtárik
Stochastic Gradient Descent Practical Guide Estimation Bias Model Zoo Unbiased Gradient Common Bias

March 27, 2023

Adjusted Wasserstein Distributionally Robust Estimator in Statistical Learning
Yiling Xie, Xiaoming Huo
Statistical Learning Robust Estimation Estimation Bias Wasserstein Distributionally Robust Wasserstein 2

January 20, 2023

Revisiting Estimation Bias in Policy Gradients for Deep Reinforcement Learning
Haoxuan Pan, Deheng Ye, Xiaoming Duan, Qiang Fu, Wei Yang, Jianping He, Mingfei Sun
Deep Reinforcement Learning Policy Gradient Policy OpTimization Estimation Bias

December 31, 2022

On High dimensional Poisson models with measurement error: hypothesis testing for nonlinear nonconvex optimization
Fei Jiang, Yeqing Zhou, Jianxuan Liu, Yanyuan Ma
Estimation Task Nonconvex Optimization Hypothesis Testing Estimation Bias Measurement Error

December 28, 2022

On the Convergence of Discounted Policy Gradient Methods
Chris Nota
Reinforcement Learning Policy Gradient Early Stage Convergence Estimation Bias Gradient Ascent

December 20, 2022

Generalized Simultaneous Perturbation-based Gradient Search with Reduced Estimator Bias
Soumen Pachal, Shalabh Bhatnagar, L. A. Prashanth
Stochastic Approximation Estimation Bias Noisy Measurement Simultaneous Perturbation Stochastic Approximation

November 11, 2022

Sensor Visibility Estimation: Metrics and Methods for Systematic Performance Evaluation and Improvement
Joachim Börger, Marc Patrick Zapf, Marat Kopytjuk, Xinrun Li 2, Claudius Gläser
NCD Method Large Relevance Improvement Metric Library Performance Evaluation Estimation Bias Brand Visibility Visibility Estimation