面向类别增量学习的基于自编码器的混合回放方法
Autoencoder-Based Hybrid Replay for Class-Incremental Learning
AI总结：
本研究针对类别增量学习中的灾难性遗忘与高内存复杂度问题，提出基于混合自编码器的混合回放策略，通过潜在空间存储样本与新类质心增量嵌入，在低内存预算下实现优于现有基线的性能。
摘要：
在类别增量学习（CIL）中，有效的增量学习策略对于缓解任务混淆和灾难性遗忘至关重要，尤其是随着任务数量$t$的增加。当前的样本回放策略具有$\mathcal{O}(t)$的内存/计算复杂度。我们提出了一种基于自编码器的混合回放（AHR）策略，该策略利用我们提出的新型混合自编码器（HAE）作为压缩器来降低大内存需求，在最坏情况下实现$\mathcal{O}(0.1 t)$的内存复杂度，同时保持$\mathcal{O}(t)$的计算复杂度，并取得了当前最优的性能。随后解码器会恢复存储在潜在空间而非原始格式的样本数据。此外，HAE同时支持判别式建模和生成式建模，分别实现分类和回放功能。HAE采用带电粒子系统能量最小化方程和斥力算法，在其潜在空间中实现新类别质心的增量嵌入和分布。我们的实验结果表明，在相同的内存/计算预算下，AHR在多个基准测试中始终优于近期的基线方法。源代码包含在补充材料中，将于论文发表后开源。
英文摘要：
In class-incremental learning (CIL), effective incremental learning strategies are essential to mitigate task confusion and catastrophic forgetting, especially as the number of tasks $t$ increases. Current exemplar replay strategies impose $\mathcal{O}(t)$ memory/compute complexities. We propose an autoencoder-based hybrid replay (AHR) strategy that leverages our new hybrid autoencoder (HAE) to function as a compressor to alleviate the requirement for large memory, achieving $\mathcal{O}(0.1 t)$ at the worst case with the computing complexity of $\mathcal{O}(t)$ while accomplishing state-of-the-art performance. The decoder later recovers the exemplar data stored in the latent space, rather than in raw format. Additionally, HAE is designed for both discriminative and generative modeling, enabling classification and replay capabilities, respectively. HAE adopts the charged particle system energy minimization equations and repulsive force algorithm for the incremental embedding and distribution of new class centroids in its latent space. Our results demonstrate that AHR consistently outperforms recent baselines across multiple benchmarks while operating with the same memory/compute budgets. The source code is included in the supplementary material and will be open-sourced upon publication.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
