Abstract:While most continual learning methods focus on mitigating forgetting and improving accuracy, they often overlook the critical aspect of network calibration, despite its importance. Neural collapse, a phenomenon where last-layer features collapse to their class means, has demonstrated advantages in continual learning by reducing feature-classifier misalignment. Few works aim to improve the calibration of continual models for more reliable predictions. Our work goes a step further by proposing a novel method that not only enhances calibration but also improves performance by reducing overconfidence, mitigating forgetting, and increasing accuracy. We introduce Sphere-Adaptive Mixup (SAMix), an adaptive mixup strategy tailored for neural collapse-based methods. SAMix adapts the mixing process to the geometric properties of feature spaces under neural collapse, ensuring more robust regularization and alignment. Experiments show that SAMix significantly boosts performance, surpassing SOTA methods in continual learning while also improving model calibration. SAMix enhances both across-task accuracy and the broader reliability of predictions, making it a promising advancement for robust continual learning systems.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2510.15751 [cs.LG] |
| (or arXiv:2510.15751v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2510.15751 arXiv-issued DOI via DataCite |
From: Trung Anh Dang [view email]
[v1]
Fri, 17 Oct 2025 15:36:46 UTC (1,466 KB)
[v2]
Wed, 22 Apr 2026 15:48:22 UTC (1,544 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。