An Equivalence between Loss Functions and Non-Uniform Sampling in Experience Replay.

Scott Fujimoto David Meger Doina Precup

Published in: NeurIPS (2020)

Keyphrases

loss function
pairwise
support vector
squared error
learning to rank
loss minimization
logistic regression
empirical risk
boosting framework
learning models
convex loss functions
hinge loss
information retrieval
reproducing kernel hilbert space
update rules