Timezone: »

Perfectly Balanced: Improving Transfer and Robustness of Supervised Contrastive Learning
Mayee Chen · Daniel Y Fu · Avanika Narayan · Michael Zhang · Zhao Song · Kayvon Fatahalian · Christopher Re

Wed Jul 20 03:30 PM -- 05:30 PM (PDT) @ Hall E #1126

An ideal learned representation should display transferability and robustness. Supervised contrastive learning (SupCon) is a promising method for training accurate models, but produces representations that do not capture these properties due to class collapse---when all points in a class map to the same representation. Recent work suggests that "spreading out" these representations improves them, but the precise mechanism is poorly understood. We argue that creating spread alone is insufficient for better representations, since spread is invariant to permutations within classes. Instead, both the correct degree of spread and a mechanism for breaking this invariance are necessary. We first prove that adding a weighted class-conditional InfoNCE loss to SupCon controls the degree of spread. Next, we study three mechanisms to break permutation invariance: using a constrained encoder, adding a class-conditional autoencoder, and using data augmentation. We show that the latter two encourage clustering of latent subclasses under more realistic conditions than the former. Using these insights, we show that adding a properly-weighted class-conditional InfoNCE loss and a class-conditional autoencoder to SupCon achieves 11.1 points of lift on coarse-to-fine transfer across 5 standard datasets and 4.7 points on worst-group robustness on 3 datasets, setting state-of-the-art on CelebA by 11.5 points.

Author Information

Mayee Chen (Stanford University)
Daniel Y Fu (Stanford University)
Avanika Narayan (Stanford University)
Michael Zhang (Stanford University)
Zhao Song (Adobe Research)
Kayvon Fatahalian (Stanford)
Christopher Re (Stanford University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors