Timezone: »

 
Poster
Not All Semantics are Created Equal: Contrastive Self-supervised Learning with Automatic Temperature Individualization
Zi-Hao Qiu · Quanqi Hu · Zhuoning Yuan · Denny Zhou · Lijun Zhang · Tianbao Yang

Thu Jul 27 01:30 PM -- 03:00 PM (PDT) @ Exhibit Hall 1 #304
in Poster Session 5 »
In this paper, we aim to optimize a contrastive loss with individualized temperatures in a principled manner. The common practice of using a global temperature parameter $\tau$ ignores the fact that ``not all semantics are created equal", meaning that different anchor data may have different numbers of samples with similar semantics, especially when data exhibits long-tails. First, we propose a new robust contrastive loss inspired by distributionally robust optimization (DRO), providing us an intuition about the effect of $\tau$ and a mechanism for automatic temperature individualization. Then, we propose an efficient stochastic algorithm for optimizing the robust contrastive loss with a provable convergence guarantee without using large mini-batch sizes. Theoretical and experimental results show that our algorithm automatically learns a suitable $\tau$ for each sample. Specifically, samples with frequent semantics use large temperatures to keep local semantic structures, while samples with rare semantics use small temperatures to induce more separable features. Our method not only outperforms prior strong baselines (e.g., SimCLR, CLIP) on unimodal and bimodal tasks with larger improvements on imbalanced data but also is less sensitive to hyper-parameters. To our best knowledge, this is the first methodical approach to optimizing a contrastive loss with individualized temperatures. Our proposed algorithms are implemented in the LibAUC library at https://libauc.org.
[ PDF]  [ Poster] 

Author Information

Zi-Hao Qiu (Nanjing University)
Quanqi Hu (Texas A&M University)
Zhuoning Yuan (The University of Iowa)

Ph.D. student in Machine Learning

Denny Zhou (Google Brain)
Lijun Zhang (Nanjing University)
Tianbao Yang (Texas A&M University)

More from the Same Authors