发表机构
Florida State University（佛罗里达州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究提出一种自监督图像聚类框架，无需明确目标。基于最大熵原理定义“惊喜分数”，采用“收敛到惊喜”方案，通过进化策略外循环与梯度下降内循环结合优化模型，在非参数自监督图像聚类中取得新的最优成果。
AI 中文摘要
大多数自监督图像聚类模型基于梯度下降，计算损失时需明确目标。本文提出无此要求的自监督框架。基于最大熵原理假设像素独立同分布作为原假设H0，定义“惊喜分数”衡量模型输出表示在H0下的不可能性。最大化惊喜分数促使模型拒绝H0以发现非随机特征。提出“收敛到惊喜”方案优化模型，由进化策略外循环直接最大化惊喜分数，与周期梯度下降内循环结合，以内循环发现的令人惊讶的聚类为替代目标。在标准图像基准测试中，该框架在非参数自监督图像聚类中取得新的最优结果。
英文摘要
A variety of self-supervised image clustering approaches are invented in the past years. However, all dominant approaches are exploitative: The direction of parameter updates is determined by known states (observed input samples and existing parameters). We propose an explorative self-supervised learning framework that steps out of this zone. We define a surprise score that measures how unlikely the model's output representation is, assuming that all pixels are i.i.d. random noise. Maximizing the surprise score forces the deep learning model to reject the random noise null hypothesis, or equivalently, to discover non-randomness from data. Also, we propose a fundamental assumption: a surprise score cannot, in general, be fully optimized by exploitative optimization approaches. Thus, we propose the converge-to-surprise scheme to optimize a model: an evolution-strategy (ES) outer loop, which maximizes the surprise score using the mutation-selection mechanism, paired with a periodic gradient-descent inner loop, which uses the surprising clusters already discovered by ES as surrogate targets. On simple image benchmarks, our framework trained from scratch achieves new state-of-the-art results in non-parametric self-supervised image clustering --- the strictest deep-clustering setting, where the number of classes is unknown during training.
