Under SETH, approximating integrals, Poisson solutions, or matrix-vector products for neural network inputs requires runtime at least accuracy^{-1+o(1)}.
Neural networks as smooth priors for inverse problems for PDEs
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Computational Math with Neural Networks is Hard
Under SETH, approximating integrals, Poisson solutions, or matrix-vector products for neural network inputs requires runtime at least accuracy^{-1+o(1)}.