2 problems

0 votes0 replies0 views
Order-of-limits conjecture for smooth neural-network asymptotics
For the neural-network model $\mathcal{F}_{\tanh}$ , let $L$ denote the depth and $D$ the dimension, and consider the limits in which both parameters grow unboundedly for the smooth…
Conjectureopenneural-network-limits neural-tangent-kernels residual-neural-networksarXiv posted 3 days ago
0 votes0 replies0 views
Joint-width-depth limit conjecture for smooth neural-network activations
Consider the doubly infinite residual-network regime in which the depth $L$ and dimension $D$ both tend to infinity, with smooth activation functions as in the paper. Joint limit c…
Conjectureopenneural-network-limits residual-neural-networks stochastic-processesarXiv posted 3 days ago

Order-of-limits conjecture for smooth neural-network asymptotics