You are here: Home Talks & Events Bernstein Seminar 2013 John Hertz (Nordita and KTH Royal Institute of Technology, Stockholm, Sweden)

John Hertz (Nordita and KTH Royal Institute of Technology, Stockholm, Sweden)

— filed under:

"Network Inference with Hidden Units" / Thursday, January 17, 2013, 17:15 h

  • Bernstein Seminar
When Jan 17, 2013
from 05:15 PM to 06:45 PM
Where Lecture Hall, Hansastr. 9a
Add event to calendar vCal









The Bernstein Center Freiburg

Bernstein Seminar
John Hertz
Nordita and KTH Royal Institute of Technology, Stockholm, Sweden

Network Inference with Hidden Units

Thursday, January 17, 2013

17:15 h

Lecture Hall (ground floor)
Bernstein Center Freiburg
Hansastraße 9A
79104 Freiburg

We derive learning rules for finding the connections between units in stochastic dynamical networks from the recorded history of a “visible” subset of the units. We consider two models. In both of them, the visible units are binary and stochastic. In one model the “hidden” units are continuous-valued, with sigmoidal activation functions, and in the other they are binary and stochastic like the visible ones. We derive exact learning rules for both cases. For the stochastic case, performing the exact calculation requires, in general, repeated summations over an number of configurations that grows exponentially with the size of the system and the data length, which is not feasible for large systems. We derive a mean field theory, based on a factorized ansatz for the distribution of hidden-unit states, which offers an attractive alternative for large systems. We present the results of some numerical calculations that illustrate key features of the two models and, for the stochastic case, the exact and approximate calculations.

The talk is open to the public. Guests are cordially invited!


Personal tools