Long-Tailed Partial Label Learning via Dynamic Rebalancing
This addresses a critical problem for machine learning practitioners dealing with real-world noisy and imbalanced data, representing an incremental improvement over prior LT-PLL approaches.
The paper tackles the combined challenge of label ambiguity and heavy class imbalance in partial label learning (PLL) under long-tailed distributions, showing that existing methods underperform even with oracle class priors. It proposes RECORDS, a dynamic rebalancing method that achieves significant gains over baselines on three benchmark datasets.
Real-world data usually couples the label ambiguity and heavy imbalance, challenging the algorithmic robustness of partial label learning (PLL) and long-tailed learning (LT). The straightforward combination of LT and PLL, i.e., LT-PLL, suffers from a fundamental dilemma: LT methods build upon a given class distribution that is unavailable in PLL, and the performance of PLL is severely influenced in long-tailed context. We show that even with the auxiliary of an oracle class prior, the state-of-the-art methods underperform due to an adverse fact that the constant rebalancing in LT is harsh to the label disambiguation in PLL. To overcome this challenge, we thus propose a dynamic rebalancing method, termed as RECORDS, without assuming any prior knowledge about the class distribution. Based on a parametric decomposition of the biased output, our method constructs a dynamic adjustment that is benign to the label disambiguation process and theoretically converges to the oracle class prior. Extensive experiments on three benchmark datasets demonstrate the significant gain of RECORDS compared with a range of baselines. The code is publicly available.