Learning to Defer to a Population: A Meta-Learning Approach

TL;DR

This paper introduces a meta-learning framework for learning to defer to unseen experts at test time, enabling autonomous systems to adapt quickly to new human or machine experts without retraining.

Contribution

It proposes a novel meta-learning approach for learning to defer that handles never-before-seen experts, using both optimization- and model-based methods with attention mechanisms.

Findings

Effective adaptation to new experts demonstrated on multiple benchmarks.

Meta-learning approach outperforms traditional methods in deferral accuracy.

Attention-based model improves expert assessment precision.

Abstract

The learning to defer (L2D) framework allows autonomous systems to be safe and robust by allocating difficult decisions to a human expert. All existing work on L2D assumes that each expert is well-identified, and if any expert were to change, the system should be re-trained. In this work, we alleviate this constraint, formulating an L2D system that can cope with never-before-seen experts at test-time. We accomplish this by using meta-learning, considering both optimization- and model-based variants. Given a small context set to characterize the currently available expert, our framework can quickly adapt its deferral policy. For the model-based approach, we employ an attention mechanism that is able to look for points in the context set that are similar to a given test point, leading to an even more precise assessment of the expert's abilities. In the experiments, we validate our methods on image recognition, traffic sign detection, and skin lesion diagnosis benchmarks.