Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces Dual-Axis Multi-Expert Learning (DAMEL) to address bias-variance trade-offs in class-imbalanced learning. DAMEL uses multiple experts along representation (concatenated representations with an auxiliary balanced classifier) and time axes (weight averaging across epochs) to reduce both bias and variance. Experiments on class-imbalanced datasets demonstrate DAMEL's effectiveness in improving prediction accuracy by mitigating both bias and variance.
Class-imbalanced learning gets a boost: DAMEL's dual-axis approach slashes both prediction bias and variance, outperforming existing methods.
Various algorithms have been proposed to address the challenges posed by class-imbalanced learning from real-world data with long-tailed distributions. While these algorithms reduce prediction bias through rebalancing techniques, they often introduce increased prediction variance as a trade-off. Several multi-expert learning algorithms aim to address this variance but involve complex procedures. We propose a new multi-expert learning algorithm, called the dual-axis multi-expert learning (DAMEL), which reduces both bias and variance of predictions by using multiple experts along both representation and time axes. Along the representation axis, DAMEL concatenates the representations of multiple experts and trains an auxiliary balanced classifier simultaneously with the concatenated representations. Along the time axis, DAMEL aggregates network weights across training epochs, employing these aggregated weights during testing. Experimental results demonstrate that DAMEL reduces both bias and variance of predictions, highlighting its effectiveness in class-imbalanced learning.