On the Effectiveness of Supervision in Asymmetric Non-Contrastive Learning
This work addresses the gap in supervised representation learning for researchers and practitioners, offering an incremental improvement over existing methods.
The paper tackles the problem of extending asymmetric non-contrastive learning (ANCL) to supervised scenarios, showing that supervised ANCL (SupSiam and SupBYOL) improves representation learning by reducing intra-class variance and achieves superior performance across various datasets and tasks.
Supervised contrastive representation learning has been shown to be effective in various transfer learning scenarios. However, while asymmetric non-contrastive learning (ANCL) often outperforms its contrastive learning counterpart in self-supervised representation learning, the extension of ANCL to supervised scenarios is less explored. To bridge the gap, we study ANCL for supervised representation learning, coined SupSiam and SupBYOL, leveraging labels in ANCL to achieve better representations. The proposed supervised ANCL framework improves representation learning while avoiding collapse. Our analysis reveals that providing supervision to ANCL reduces intra-class variance, and the contribution of supervision should be adjusted to achieve the best performance. Experiments demonstrate the superiority of supervised ANCL across various datasets and tasks. The code is available at: https://github.com/JH-Oh-23/Sup-ANCL.