Generalization error of linearized neural networks: staircase and double-descent

Neyman Seminar
Neyman Seminar
Probability Seminar
Feb 12, 2020 4:00pm to 5:00pm
1011 Evans Hall
Happening As Scheduled
Deep learning methods operate in regimes that defy the traditional statistical mindset. Despite the non-convexity of empirical risks and the huge complexity of neural network architectures, stochastic gradient algorithms can often find the global minimizer of the training loss and achieve small generalization error on test data. As one possible explanation to the training efficiency of neural...
Song Mei, Stanford University