CVLGJun 10, 2022

Balanced Product of Calibrated Experts for Long-Tailed Recognition

arXiv:2206.05260v343 citationsh-index: 55Has Code
Originality Incremental advance
AI Analysis

This work addresses the problem of distribution shift in long-tailed recognition for computer vision applications, representing an incremental improvement over prior ensemble methods.

The paper tackles the challenge of long-tailed recognition by proposing a Balanced Product of Experts (BalPoE) method, which achieves new state-of-the-art results on datasets like CIFAR-100-LT, ImageNet-LT, and iNaturalist-2018.

Many real-world recognition problems are characterized by long-tailed label distributions. These distributions make representation learning highly challenging due to limited generalization over the tail classes. If the test distribution differs from the training distribution, e.g. uniform versus long-tailed, the problem of the distribution shift needs to be addressed. A recent line of work proposes learning multiple diverse experts to tackle this issue. Ensemble diversity is encouraged by various techniques, e.g. by specializing different experts in the head and the tail classes. In this work, we take an analytical approach and extend the notion of logit adjustment to ensembles to form a Balanced Product of Experts (BalPoE). BalPoE combines a family of experts with different test-time target distributions, generalizing several previous approaches. We show how to properly define these distributions and combine the experts in order to achieve unbiased predictions, by proving that the ensemble is Fisher-consistent for minimizing the balanced error. Our theoretical analysis shows that our balanced ensemble requires calibrated experts, which we achieve in practice using mixup. We conduct extensive experiments and our method obtains new state-of-the-art results on three long-tailed datasets: CIFAR-100-LT, ImageNet-LT, and iNaturalist-2018. Our code is available at https://github.com/emasa/BalPoE-CalibratedLT.

Code Implementations2 repos
Foundations

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

Your Notes