解决原型自监督学习中原型坍缩问题，提升表征多样性与性能。
Why Prototypes Collapse: Diagnosing and Preventing Partial Collapse in Prototypical Self-Supervised Learning
- 分离编码器与原型的优化目标，避免联合训练引发的冗余
- 使用在线EM风格的高斯混合模型更新原型，保持多样性
- 无需额外正则化，显著提升下游任务表现
原型自监督学习方法普遍存在部分原型坍缩现象，即多个原型收敛到几乎相同的表示。这削弱了其核心目标——提供多样且信息丰富的目标以引导编码器学习丰富表征，导致实践者不得不过度参数化原型集或添加临时正则化，仅缓解症状而非根治。我们通过实证分析发现，坍缩源于编码器与原型的联合优化，促使一种捷径学习：训练初期原型趋向于最小化损失的冗余表示，而非提升表征多样性。为此，我们提出完全解耦的训练策略，分别在独立目标下学习编码器与原型。具体地，将原型建模为通过在线EM式过程更新的高斯混合模型，与编码器损失无关。该简单而严谨的解耦方法无需显式正则化即可消除原型坍缩，始终生成多样化原型，并带来更强的下游性能。
原文摘要 · Abstract (English)
Prototypical self-supervised learning methods consistently suffer from partial prototype collapse, where multiple prototypes converge to nearly identical representations. This undermines their central purpose -- providing diverse and informative targets to guide encoders toward rich representations -- and has led practitioners to over-parameterize prototype sets or add ad-hoc regularizers, which mitigate symptoms rather than address the root cause. We empirically trace the collapse to the joint optimization of encoders and prototypes, which encourages a type of shortcut learning: early in training prototypes drift toward redundant representations that minimize loss without necessarily enhancing representation diversity. To break the joint optimization, we introduce a fully decoupled training strategy that learns prototypes and encoders under separate objectives. Concretely, we model prototypes as a Gaussian mixture updated with an online EM-style procedure, independent of the encoder's loss. This simple yet principled decoupling eliminates prototype collapse without explicit regularization and yields consistently diverse prototypes and stronger downstream performance.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。