Landscape analysis for shallow neural networks: Complete classification of critical points for cubic activation and affine target functions
In this paper, we study the optimization landscape induced by the true loss for shallow polynomial neural networks (PNNs) with $\mathfrak{h} \in \mathbb{N}$ neurons on the hidden layer, one-dimensional input and output layers, and a monomial activation of degree $d \in \mathbb{N}$, trained against a non-constant affine linear target function. Our first main result provides for arbitrary activation degree $d$ a sharp existence/non-existence criterion for \emph{global minimizers} with necessary structural conditions. We show that the infimum of the loss is always zero and achievable with at…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2607.15173
curl -X POST /api/v1/records/arxiv/2607.15173/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2607.15173/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'