Not Eliminate but Aggregate: Post-Hoc Control over Mixture-of-Experts to Address Shortcut Shifts in Natural Language Understanding

On Universally Optimal Algorithms for A/B Testing

機械学習が紡ぐゲーム理論のフロンティア

Scalable and Provably Fair Exposure Control for Large-Scale Recommender Systems

On Uniformly Optimal Algorithms for Best Arm Identification in Two-Armed Bandits with Fixed Budget

Rate-Optimal Bayesian Simple Regret in Best Arm Identification

An Optimal Clustering Algorithm for the Labeled Stochastic Block Model

iMixer: invertible, implicit, and iterative MLP-Mixer from modern Hopfield networks

Fairness Concepts for Indivisible Items with Externalities

Unified Perspective on Probability Divergence via the Density-Ratio Likelihood: Bridging KL-Divergence and Integral Probability Metrics

公平性を考慮した大規模推薦システム

Thresholded Lasso Bandit

Learning Causal Relationship from Conditional Moment Condition by Importance Weighting

Benign-Overfitting in Conditional Average Treatment Effect Prediction with Linear Regression

Mean Variance Efficient Reinforcement Learning

Online-to-offline advertisements as field experiments

A Real-World Implementation of UnbiasedLift-based Bidding System

The Adaptive Doubly Robust Estimator and a Paradox Concerning Logging Policy

Non-Negative Bregman Divergence Minimization for Deep Direct Density Ratio Estimation

広告配信オークションにおける入札戦略

Detecting multi-timescale consumption patterns from receipt data: A non-negative tensor factorization approach