CLCE: An Approach to Refining Cross-Entropy and Contrastive Learning for Optimized Learning Fusion

arXiv:2402.14551v1 Announce Type: cross
Abstract: State-of-the-art pre-trained image models predominantly adopt a two-stage approach: initial unsupervised pre-training on large-scale datasets followed by task-specific fine-tuning using Cross-Entropy loss~(CE). However, it has been demonstrated that CE can compromise model generalization and stability. While recent works employing contrastive learning address some of these limitations by enhancing the quality of embeddings and producing better decision boundaries, they often overlook the importance of hard negative mining and rely on resource intensive and slow training using …

abstract art arxiv cross-entropy cs.ai cs.cv cs.lg datasets entropy fine-tuning fusion image loss model generalization pre-training scale stability stage state training type unsupervised