Principled Approaches for Learning to Defer with Multiple Experts
This work addresses the challenge of improving decision-making systems by deferring to multiple experts, though it is incremental as it builds on existing learning-to-defer frameworks.
The paper tackles the problem of learning to defer predictions to multiple experts by introducing a new family of surrogate losses and proving strong H-consistency bounds for them, with experimental results reported on SVHN and CIFAR-10 datasets.
We present a study of surrogate losses and algorithms for the general problem of learning to defer with multiple experts. We first introduce a new family of surrogate losses specifically tailored for the multiple-expert setting, where the prediction and deferral functions are learned simultaneously. We then prove that these surrogate losses benefit from strong $H$-consistency bounds. We illustrate the application of our analysis through several examples of practical surrogate losses, for which we give explicit guarantees. These loss functions readily lead to the design of new learning to defer algorithms based on their minimization. While the main focus of this work is a theoretical analysis, we also report the results of several experiments on SVHN and CIFAR-10 datasets.