Improving Domain Generalization in Contrastive Learning using Adaptive Temperature Control

TL;DR

This paper introduces an adaptive temperature control method in contrastive learning that leverages domain labels to enhance domain invariance and improve out-of-distribution generalization, especially under significant covariate shifts.

Contribution

The paper proposes a novel contrastive learning approach that dynamically adjusts the temperature parameter based on domain similarity, improving domain generalization in unseen environments.

Findings

Outperforms baseline methods on out-of-distribution tests.

Maintains strong in-distribution task performance.

Effective in a multi-domain MNIST variant.

Abstract

Self-supervised pre-training with contrastive learning is a powerful method for learning from sparsely labeled data. However, performance can drop considerably when there is a shift in the distribution of data from training to test time. We study this phenomenon in a setting in which the training data come from multiple domains, and the test data come from a domain not seen at training that is subject to significant covariate shift. We present a new method for contrastive learning that incorporates domain labels to increase the domain invariance of learned representations, leading to improved out-of-distribution generalization. Our method adjusts the temperature parameter in the InfoNCE loss -- which controls the relative weighting of negative pairs -- using the probability that a negative sample comes from the same domain as the anchor. This upweights pairs from more similar domains, encouraging the model to discriminate samples based on domain-invariant attributes. Through experiments on a variant of the MNIST dataset, we demonstrate that our method yields better out-of-distribution performance than domain generalization baselines. Furthermore, our method maintains strong in-distribution task performance, substantially outperforming baselines on this measure.