3 problems

0 votes0 replies0 views
Width independence conjecture for residual networks
Width independence conjecture. We conjecture that $m$ is independent of $L$ .
Conjectureopendeep-learning neural-tangent-kernel residual-networksarXiv posted 2 days ago
0 votes0 replies1 view
Conjecture on two-dimensional loss landscapes of wide neural networks
Let $F$ be the loss of a neural network and let $\theta^*$ be a global minimum. Choose two vectors $v_1,v_2$ according to a specified rule, for example a Gaussian distribution, and…
Conjectureopenneural-networks optimization-landscapes residual-networksarXiv posted 2 days ago
0 votes0 replies0 views
The residual-network output-preservation conjecture
Output-preservation conjecture. As long as $\hat{\mathbf y}\approx {\mathbf y}$ , namely
Conjectureopenneural-networks quantization residual-networksarXiv posted 2 days ago

Width independence conjecture for residual networks