A Conjugate Property between Loss Functions and Uncertainty Sets in Classification Problems.

Takafumi Kanamori Akiko Takeda Taiji Suzuki

Published in: COLT (2012)

Keyphrases

loss function
support vector
pairwise
squared error
loss minimization
logistic regression
learning to rank
empirical risk
risk minimization
boosting framework
learning models
reproducing kernel hilbert space