NTK-Guided Few-Shot Class Incremental Learning

arXiv:2403.12486v1 Announce Type: new
Abstract: While anti-amnesia FSCIL learners often excel in incremental sessions, they tend to prioritize mitigating knowledge attrition over harnessing the model's potential for knowledge acquisition. In this paper, we delve into the foundations of model generalization in FSCIL through the lens of the Neural Tangent Kernel (NTK). Our primary design focus revolves around ensuring optimal NTK convergence and NTK-related generalization error, serving as the theoretical bedrock for exceptional generalization. To attain globally optimal NTK convergence, we …

abstract acquisition arxiv attrition class cs.ai cs.lg design excel few-shot focus incremental kernel knowledge knowledge acquisition model generalization paper through type