no code implementations • 9 Aug 2021 • Miguel Ruiz-Garcia
This work emphasizes the importance that neural network architecture has for the emergence of catastrophic forgetting and introduces a neural network that is able to learn an algorithm.
1 code implementation • 7 Feb 2021 • Miguel Ruiz-Garcia, Ge Zhang, Samuel S. Schoenholz, Andrea J. Liu
In underparameterized networks, such dynamical loss functions can lead to successful training for networks that fail to find a deep minima of the standard cross-entropy loss.