1 problem
- 0 votes0 replies0 views
The hyper-parameter conjecture for Regularized Top- sparsification
Let denote the accumulated gradient associated with coordinate at iteration , and let be the exponent governing the prior's dependence on…
The hyper-parameter conjecture for Regularized Top- sparsification
Let denote the accumulated gradient associated with coordinate at iteration , and let be the exponent governing the prior's dependence on…