Certifying Residual Architectures from Their Primitives: A Sharp Stability Threshold

Excerpt

arXiv:2607.14576v2 Announce Type: replace Abstract: Whether a deep residual architecture trains stably is usually determined by training it, which is expensive and answers the question only for the architecture, depth, and floating-point format tested. In practice, stability is secured by heuristics for where to place normalization and which type to use, supported by experiments but lacking a common principle. We show that stability can instead be certified before training, directly from the arc