The generic total H-complexity conjecture for ReLU neural network maps
Let be a ReLU neural network map with canonical polyhedral decomposition . The total -complexity of and the -cells of are understood as defined in the paper.
Generic total H-complexity conjecture. With probability , the total -complexity of is less than or equal to the number of -cells of .
The conjecture is motivated by the contrast between high-complexity PL examples with a single -cell and the claim that such behavior is highly atypical for ReLU neural network functions. The supplied text gives no resolution.
References
- J. Elisenda Grigsby, Kathryn Lindsey and Marissa Masden, “Local and global topological complexity measures OF ReLU neural network functions”, arXiv:2204.06062 (2024).
Progress summary
Never refreshed
Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.
Solutions 0
No solutions have been posted yet.