MLAILGNov 28, 2024

Concept-driven Off Policy Evaluation

arXiv:2411.19395v1h-index: 3
Originality Incremental advance
AI Analysis

This work addresses variance reduction in OPE for reinforcement learning applications, offering an interpretable solution that is incremental over existing methods.

The paper tackles the problem of high variance in Off-Policy Evaluation (OPE) with limited batch data by incorporating human-explainable concepts into estimators, proving they remain unbiased and reduce variance when concepts are known, and developing an algorithm to learn optimized concepts when they are not. Experiments on synthetic and real-world datasets show significant performance improvements, with the estimators also being interpretable and allowing targeted interventions.

Evaluating off-policy decisions using batch data poses significant challenges due to limited sample sizes leading to high variance. To improve Off-Policy Evaluation (OPE), we must identify and address the sources of this variance. Recent research on Concept Bottleneck Models (CBMs) shows that using human-explainable concepts can improve predictions and provide better understanding. We propose incorporating concepts into OPE to reduce variance. Our work introduces a family of concept-based OPE estimators, proving that they remain unbiased and reduce variance when concepts are known and predefined. Since real-world applications often lack predefined concepts, we further develop an end-to-end algorithm to learn interpretable, concise, and diverse parameterized concepts optimized for variance reduction. Our experiments with synthetic and real-world datasets show that both known and learned concept-based estimators significantly improve OPE performance. Crucially, we show that, unlike other OPE methods, concept-based estimators are easily interpretable and allow for targeted interventions on specific concepts, further enhancing the quality of these estimators.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes