[Ml-stat-talks] Fwd: [ORFE-Seminars] Wilks Statistics Seminar: Francesco Orabona, Stony Brook University: Monday, November 13 at 12:30pm, Sherrerd Hall 101

Barbara Engelhardt bee at princeton.edu
Tue Nov 7 17:59:06 EST 2017

Talk of interest next Monday.

****   Wilks Statistics Seminar   *** *

*DATE:                 Monday, November 13, 2017 TIME:
12:30 pm LOCATION:       Sherrerd Hall 101 SPEAKER:         Francesco
Orabona, Stony Brook University TITLE:               **Parameter-free
Machine Learning through Coin Betting*

  *Abstract:*  Machine Learning (ML) has been described as the fuel of the
next industrial revolution. Yet, despite their name, the majority of the ML
algorithms still heavily rely on having humans in the loop in order to set
their "parameters". For example, when using regularized empirical risk
minimization, the choice of the weight of the regularizer is critical to
obtain theoretical and practical optimal performance. Moreover, the
minimization itself, usually done through stochastic gradient descent
procedures, requires to set "step sizes" in order to get good performance.
Are these parameters strictly necessary? Is it possible to have
parameter-free ML algorithms? In this talk, I will show that both the
problems of ML with convex Lipschitz losses and in general stochastic
optimization of convex Lipschitz functions can be reduced to a game of
betting on a non-stochastic coin. Betting on a non-stochastic coin is a
well known problem whose optimal strategy turns out to be a simple
generalization of the Kelly betting criterion. Moreover, the optimal coin
betting algorithm is parameter-free, giving rise to parameter-free ML and
stochastic optimization algorithms. In particular, this approach gives
optimal rates of convergence in RKHS in the capacity independent setting
without any parameter to tune. Empirically results will be shown as well.

* ********************************* **Short Biography:  *Francesco Orabona
is an Assistant Professor at Stony Brook University. His research interests
are in the area of theoretically motivated and efficient machine learning
algorithms, with emphasis on online and stochastic methods. He received the
PhD degree in Electrical Engineering at the University of Genoa, in 2007.
He is (co)author of more than 60 peer reviewed papers.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/ml-stat-talks/attachments/20171107/86c0246c/attachment.html>

More information about the Ml-stat-talks mailing list