Generalization Beyond Feature Alignment: Concept Activation-Guided Contrastive Learning
원문 초록
Learning invariant representations via contrastive learning has seen state-of-the-art performance in domain generalization (DG). Despite such success, in this paper, we find that its core learning strategy -- feature alignment -- could heavily hinder model generalization. Drawing insights in neuron interpretability, we characterize this problem from a neuron activation view. Specifically, by treating feature elements as neuron activation states, we show that conventional alignment methods tend to deteriorate the diversity of learned invariant features, as they indiscriminately minimize all neuron activation differences. This instead ignores rich relations among neurons -- many of them often identify the same visual concepts despite differing activation patterns. With this finding, we present a simple yet effective approach, Concept Contrast (CoCo), which relaxes element-wise feature alignments by contrasting high-level concepts encoded in neurons. Our CoCo performs in a plug-and-play fashion, thus it can be integrated into any contrastive method in DG. We evaluate CoCo over four canonical contrastive methods, showing that CoCo promotes the diversity of feature representations and consistently improves model generalization capability. By decoupling this success through neuron coverage analysis, we further find that CoCo potentially invokes more meaningful neurons during training, thereby improving model learning.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Prompt-Free Conditional Diffusion for Multi-object Image Augmentation
FixCLR: Negative-Class Contrastive Learning for Semi-Supervised Domain Generalization
Controllable Expressive 3D Facial Animation via Diffusion in a Unified Multimodal Space
Avoiding Shortcuts: Enhancing Channel-Robust Specific Emitter Identification via Single-Source Domain Generalization
SSCL-IDS: Enhancing Generalization of Intrusion Detection with Self-Supervised Contrastive Learning
Instance Paradigm Contrastive Learning for Domain Generalization
토론
이 논문과 연결된 포럼 글이 아직 없습니다.