[Ml-stat-talks] jake abernethy
Robert Schapire
schapire at CS.Princeton.EDU
Mon Apr 12 10:37:38 EDT 2010
Jake Abernethy from UC Berkeley is speaking TODAY (Monday) at 4:30PM in
the Computer Science Small Auditorium (Room 105).
Rob
*Learning, Adversaries, and Limited Feedback*
*Jake Abernethy*
UC Berkeley
A basic assumption often made in Machine Learning is that the data
we observe are independent and identically distributed. But is this
really necessary? Is it even realistic in typical scenarios? A
number of recent results provide guarantees for arbitrary (or even
adversarially-generated) data sequences. It turns out, for a wide
class of problems, the learner's performance is no worse when the
data is assumed to be arbitrary as opposed to i.i.d. This result
involves a nice application of the Minimax theorem, which I'll
briefly describe. I'll also delve into a more challenging problem:
the "bandit" setting, in which the learner receives only very little
feedback on each example. It was unknown for some time whether there
exists an efficient algorithm that achieves the same guarantee for
non-i.i.d. data for a general class of learning/decision problems.
I'll present the first known such bandit algorithm, and I'll sketch
the central ideas behind the technique, borrowing several tricks
from the Interior Point optimization literature.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/ml-stat-talks/attachments/20100412/806ea613/attachment.htm>
More information about the Ml-stat-talks
mailing list