Locally Adaptive Multi-Objective Learning
This addresses the challenge of multi-objective learning in non-stationary environments for applications like energy forecasting and algorithmic fairness, representing an incremental improvement over prior work.
The paper tackles the problem of learning predictors that satisfy multiple objectives under arbitrary distribution shifts over time, proposing a method that achieves local adaptivity by incorporating an adaptive online algorithm. Empirical results on energy forecasting and algorithmic fairness datasets show the method improves upon existing approaches and achieves unbiased predictions across subgroups while maintaining robustness to distribution shifts.
We consider the general problem of learning a predictor that satisfies multiple objectives of interest simultaneously, a broad framework that captures a range of specific learning goals including calibration, regret, and multiaccuracy. We work in an online setting where the data distribution can change arbitrarily over time. Existing approaches to this problem aim to minimize the set of objectives over the entire time horizon in a worst-case sense, and in practice they do not necessarily adapt to distribution shifts. Earlier work has aimed to alleviate this problem by incorporating additional objectives that target local guarantees over contiguous subintervals. Empirical evaluation of these proposals is, however, scarce. In this article, we consider an alternative procedure that achieves local adaptivity by replacing one part of the multi-objective learning method with an adaptive online algorithm. Empirical evaluations on datasets from energy forecasting and algorithmic fairness show that our proposed method improves upon existing approaches and achieves unbiased predictions over subgroups, while remaining robust under distribution shift.