Unverified50% confidenceFactExact time
InfoNCE损失的理论根源可追溯至DeepMind的van den Oord等人于2018年提出的CPC（Contrastive Predictive Coding）
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedInfoNCE损失最早由van den Oord等人于2018年在时序数据表示学习中提出80% similarUnverifiedThe core loss functions of contrastive learning (such as InfoNCE and NT-Xent) essentially pull the representations of similar samples closer together while pushing the representations of dissimilar samples further apart.70% similarUnverified状态离散化存在权衡：分桶粒度过粗导致信息损失，过细则引发维度灾难，这也是DQN用神经网络替代Q表的核心动机68% similarUnverifiedA reasonable prediction loss combined with 0% downstream control accuracy is highly consistent with representation collapse, meaning the encoder has fooled the self-supervised objective without learning any useful dynamic information.63% similarUnverifiedFocal Loss was proposed by Kaiming He et al. in the 2017 RetinaNet paper to address the problem of extreme foreground-background sample imbalance in object detection.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/524279API
curl https://kongchang.com/api/v1/knowledge/claims/524279MCP
get_claim(id=524279)