# The Standard Deviation in VICReg Buys a Safety Margin

Preprint · Hawar Dizayee

- Archive id: arxiv:2610.32630
- Published: 2026-10-23
- arXiv: https://arxiv.org/abs/2610.32630
- Source URL: https://arxiv.org/abs/2610.32630
- PDF: https://arxiv.org/pdf/2610.32630
- Provenance: arxiv

## Abstract

VICReg prevents collapse with a hinge on the standard deviation of each embedding dimension, and its authors argue from the gradient that a hinge on the variance would fail. To our knowledge, this claim has not been tested experimentally. We test it in a controlled setting (ResNet-18, CIFAR-10 and CIFAR-100), changing only the hinge and sweeping its weight $μ$ over five orders of magnitude. At the default weight ($μ= 1$), the two hinges reach the same linear-probe accuracy on CIFAR-10 (about 89.5%). Both collapse below a threshold weight, but the standard-deviation hinge's threshold is 35-130$\\times$ lower. CIFAR-100 shows a consistent gap. In training, the variance hinge's gradient fades as the feature variance shrinks, while the standard-deviation hinge's does not until that variance is very small. The standard-deviation hinge therefore buys a safety margin: it keeps training at weights far below the point where the variance hinge collapses.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.