Competing with the Empirical Risk Minimizer in a Single Pass thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Competing with the Empirical Risk Minimizer in a Single Pass

Published on Aug 20, 20152073 Views

Many optimization problems that arise in science and engineering are those in which we only have a stochastic approximation to the underlying objective (e.g. estimation problems such as linear regress

Related categories

Chapter list

Competing with the empirical risk minimizer in a single pass00:00
Tradeoffs in large-scale learning00:08
Approaches00:44
Competing with the ERM01:47
Main result03:01
Thanks!03:50