Wide neural networks with general weights: convergence rate and explicit dependence on the hyper-parameters
Using Stein's method techniques introduced by Chatterjee (2008) and further extended by Kasprzak and Peccati (2022) and by Lachièze-Rey and Peccati (2017), we derive novel quantitative bounds on the convergence in distribution of feed-forward fully connected neural networks (with Lipschitz activation functions) towards Gaussian processes, as the hidden layer width $n$ tends to infinity. We consider networks initialized with independent and identically distributed (i.i.d.) weights possessing sufficiently many finite moments, and i.i.d. Gaussian biases independent of the weights. Specifically,…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2601.21539
curl -X POST /api/v1/records/arxiv/2601.21539/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2601.21539/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'