Online Learning and Online Convex Optimisation
Add to your list(s)
Download to your calendar using vCal
If you have a question about this talk, please contact Elre Oldewage.
In online learning, data arrives sequentially, and model parameters are
updated at each step. This is in contrast to batch training, where all
data is available at once. Recently, online learning has large-scale
applications such as online web ranking and online advertisement
placement, and is closely related to continual learning. The field of
online learning itself is well-established, with a lot of theory.
We will closely follow “Online Learning and Online Convex Optimisation”
by Shalev-Shwartz (2011) (up to and including Section 2.5). We will see
how important convexity is, and analyse the regret of some well-known
algorithms such as Follow-The-Leader, Follow-The-Regularised-Leader, and
Online Gradient Descent.
This talk is part of the Machine Learning Reading Group @ CUED series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
|