82039d16dce0aab3913b6a7ac73deff7-AuthorFeedback.pdf
–Neural Information Processing Systems
The usage of entropy (H(si|sj) and H(ri|sj,a)) allows us to discourage such case while11 |si si sj|cannot.12 Explanation of Ldiv2: The usage of variance to approximate entropy was discussed in L203.
Neural Information Processing Systems
Feb-9-2026, 04:15:21 GMT