A Flatness-Generalization Relation in the Teacher-Student Tree-Committee Machine
The flatness of the loss landscape at a minimizer is a widely used heuristic for reasoning about neural-network generalization, yet evidence for this relation is mostly empirical and controversial, and it is found that the answer strongly depends on the learning task and on the ratio of the number of parameters to the...
B. Annesi, Davide Straziota, E. M. Malatesta
· 0 citations