353 citations · 1.2k across the 32 of their papers we have counts for
Showing 2020 · cs.LGShow all
2 papers · 2 filters
cs.LG2020★ 13 cited
Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization
Stanislaw Jastrzebski, Devansh Arpit, Oliver Astrand +6
The early phase of training a deep neural network has a dramatic effect on the local curvature of the loss function. For instance, using a small learning rate does not guarantee st…
cs.LG2020★ 32 cited
The Break-Even Point on Optimization Trajectories of Deep Neural Networks
Stanislaw Jastrzebski, Maciej Szymczak, Stanislav Fort +4
The early phase of training of deep neural networks is critical for their final performance. In this work, we study how the hyperparameters of stochastic gradient descent (SGD) use…