Aaron Roth

Amazon Scholar@Amazon·美国·OpenReview

研究方向

Algorithmic fairness · fairness · multicalibration · Differential privacy · private synthetic data · Online learning · no regret · online calibration · Algorithmic game theory · mechanism design · learning in games

7.3

Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents

ICML 2025Spotlight

三作

6.7

Auto-GDA: Automatic Domain Adaptation for Efficient Grounding Verification in Retrieval-Augmented Generation

ICLR 2025Poster

6.0

Conformal Language Model Reasoning with Coherent Factuality

ICLR 2025Poster

5.5

Intersectional Fairness in Reinforcement Learning with Large State and Constraint Spaces

ICML 2025Poster

4.9

The Relationship Between No-Regret Learning and Online Conformal Prediction

合作者 (20)

Aaron Roth

Robust Decision-Making with Partially Calibrated Forecasters

Emergent Alignment Via Competition

Replicable Reinforcement Learning with Linear Function Approximation

Dynamic Regret Bounds for Online Omniprediction with Long Term Constraints

High-Dimensional Prediction for Sequential Decision Making

Stronger Neyman Regret Guarantees for Adaptive Experimental Design

Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents

Auto-GDA: Automatic Domain Adaptation for Efficient Grounding Verification in Retrieval-Augmented Generation

Conformal Language Model Reasoning with Coherent Factuality

Intersectional Fairness in Reinforcement Learning with Large State and Constraint Spaces

The Relationship Between No-Regret Learning and Online Conformal Prediction