Sparse-Aware Neural Networks for Nonlinear Functionals: Mitigating the Exponential Dependence on Dimension

Jianfei Li, Shuo Huang, Han Feng, Ding-Xuan Zhou, Gitta Kutyniok

arXivarxiv/2604.06774Apr 8, 2026cs.LGscore 0

Deep neural networks have emerged as powerful tools for learning operators defined over infinite-dimensional function spaces. However, existing theories frequently encounter difficulties related to dimensionality and limited interpretability. This work investigates how sparsity can help address these challenges in functional learning, a central ingredient in operator learning. We propose a framework that employs convolutional architectures to extract sparse features from a finite number of samples, together with deep fully connected networks to effectively approximate nonlinear functionals.…

SourceJSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2604.06774
curl -X POST /api/v1/records/arxiv/2604.06774/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2604.06774/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'