[Colloquium] REMINDER: 4/21 Talks at TTIC: Yuhuai Wu, University of TORONTO

[Mary Marre]

*When:  *   Thursday, April 21st at 11:00 am

*Where:*    TTIC, 6045 S Kenwood Avenue, 5th Floor, Room 526

*Who:   *   Yuhuai Wu, *University of Toronto*

*Title*:    Architectural Complexity Measures of Recurrent Neural Networks

*Abstract*:
In this talk, I will present a rigorous graph-theoretic framework
describing the connecting architectures of RNNs in general. Based on the
framework, I will discuss three proposed architecture complexity measures
of RNNs: (a) the recurrent depth, which captures the RNN's over-time
nonlinear complexity, (b) the feedforward depth, which captures the local
input-output nonlinearity (similar to the "depth" in feedforward neural
networks (FNNs)), and (c) the recurrent skip coefficient which captures how
rapidly the information propagates over time. The experimental results show
that RNNs might benefit from larger recurrent depth and feedforward depth,
and increasing recurrent skip coefficient also offers performance boosts on
long term dependency problems.


Host: Nathan Srebro, nati at ttic.edu



Mary C. Marre
Administrative Assistant
*Toyota Technological Institute*
*6045 S. Kenwood Avenue*
*Room 504*
*Chicago, IL  60637*
*p:(773) 834-1757*
*f: (773) 357-6970*
*mmarre at ttic.edu <mmarre at ttic.edu>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20160421/bbee0780/attachment.htm