Skip to content

Conservative classifiers do consistently well with improving agents: characterizing statistical and online learning

· 0 citations · 52 references

TL;DR

An asymmetric variant of minimally consistent concept classes is introduced and used to provide an exact characterization of proper learning with improvements in the real-izable setting, and positive results for more natural Euclidean ball improvement sets are given.

View source

Similar papers

Jun 2026

Behavior Cloning is Not All You Need: The Optimality of On-Policy Distillation for Noisy Expert Feedback

A noisy expert model is proposed to explain why OPD can outperform SFT when training language models from imperfect teachers, and it is proved that online interaction with the noisy expert via a novel variant of OPD enables polynomial dependence on the horizon in general.

V. Sriraman, Peihan Liu, Daniel Hsu et al. · 0 citations
Book Open access Aug 2026

The Price of Fairness in Active Learning: Fundamental Limits and Optimal Label Acquisition

When a classifier is trained by actively choosing which examples to label, should its labeling budget depend on which notion of fairness it must satisfy? We answer this with the first information-theoretic lower bounds for group-fair active learning, and the answer is a sharp hierarchy across the three standard notions. Demographic Parity (DP), which equalizes the positive prediction rate across groups, can be verified from unlabeled data alone, so it adds no labels beyond ordinary learning. Equal Opportunity (EO), which equalizes each group's true positive rate, is fundamentally harder: estimating a rate conditioned on the true label forces the learner to collect labeled positives from every group, an irreducible cost that no query strategy can avoid. Equalized Odds (EqOdds), which additionally equalizes false positive rates, costs no more than EO whenever positives are the minority, the typical regime. This exposes a separation at the heart of fair learning: active querying yields the well-known exponential savings for accuracy, but no speedup whatsoever for fairness verification. Concretely, certifying EO to fairness tolerance γ requires Ω(d/ε2 + k/(ε2 p+)) labels, where d is the model's Vapnik--Chervonenkis (VC) dimension, k the number of groups, ε the target error, and p+ = Pr[Y=1] the positive rate; the fairness term is independent of ε, so it is a floor that active learning cannot lower. We give matching algorithms (ConstrainedERM-DP for DP, FairStratified for EO and EqOdds) that are tight up to logarithmic factors. The cost is largest exactly when positives are rare: at p+ = 0.01, as in fraud detection, EO demands roughly 90× more labels than DP. Experiments on Folktables, COMPAS, and Adult confirm the predicted scaling and yield 36 to 39% label savings over passive fair-learning baselines.

Chang Lu, Yizheng Zhao · 0 citations
Preprint Jul 2026

Weak-to-Strong Learning in Decision Making

A decision-aware weak-to-strong (W2S) framework that leverages both labeled and unlabeled data to improve contextual stochastic optimization and establishes a non-asymptotic upper bound on the excess decision risk of W2S and a complementary lower bound for a strong-only benchmark.

Jingwei Ji, Renyuan Xu · 0 citations
Review Aug 2026

On the Structural Limits of Machine Learning Decision Systems: An Information-Theoretic, Interaction-Based, and Stochastic-Dynamical Perspective

This work examines intrinsic limits of data-driven decision systems from an information-theoretic and interaction-based perspective and describes decision systems, including LLM-integrated agent architectures, as feedback-driven stochastic processes where state-dependent dynamics may induce emergent macroscopic behavior.

N. R. Barraza, G. Pena · 0 citations
Preprint Jul 2026

Post-Learning Inference for Combinatorial Optimizers with High-Dimensional Sparse Contextual Information via Minimal Directional Perturbation

A novel perturbation test based on a nonsmooth max-difference revenue statistic comparing the best null assortment with the best alternative assortment and asymptotic validity of the proposed p-value under adaptive assortment selection is proposed.

Pengyu Li, Shuting Shen · 0 citations