Methods › General › Loss Functions › DHEL
Decoupled Hyperspherical Energy Loss
DHEL
Introduced by Panagiotis Koromilas et al. in Bridging Mini-Batch and Asymptotic Analysis in Contrastive Learning: From InfoNCE to Kernel-Based Losses
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
InfoNCE variants demonstrate direct and indirect coupling between the alignment and uniformity terms thus hurting optimisation. The Decoupled Hyperspherical Energy Loss (DHEL) is an NT-Xent variant that completly decouples alignment from uniformity by discarding the corresponding terms from the denominator. In this way optimisation is more efficient and robust to hyper parameter changes.
Papers archive 2025-07-28
1 shown of 1, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
Bridging Mini-Batch and Asymptotic Analysis in Contrastive Learning: From InfoNCE to Kernel-Based Losses 28 May 2024 · 1 repository · arXiv:2405.18045Syntology ran 9 of 9 samples · 0 unverified · 9 pointer-only (licence)
Tasks archive 2025-07-28
2 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| Contrastive Learning | 1 |
| Representation Learning | 1 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections