Learning in boltzmann trees

Authors:
Lawrence Saul;Michael I. Jordan
Affiliations:
Department of Physics, Massachusetts Institute of Technology, Cambridge, MA 02139 USA;Department of Brain and Cognitive Sciences, Massachusetts institute of Technology, Cambridge, MA 02139 USA
Venue:
Neural Computation
Year:
1994

Citing 8
Cited 3

Fusion, propagation, and structuring in belief networks

Artificial Intelligence
Learning symmetry groups with hidden units: beyond the perceptron

Physica D
Probabilistic reasoning in intelligent systems: networks of plausible inference

Probabilistic reasoning in intelligent systems: networks of plausible inference
Deterministic Boltzmann learning performs steepest descent in weight-space

Neural Computation
The upstart algorithm: a method for constructing and training feedforward neural networks

Neural Computation
The Boltzmann perceptron network: a multi-layered feed-forward network equivalent to the Boltzmann machine

Advances in neural information processing systems 1
Introduction to the theory of neural computation

Introduction to the theory of neural computation
Original Contribution: A scaled conjugate gradient algorithm for fast supervised learning

Neural Networks

Distributed probabilistic inferencing in sensor networks using variational approximation

Journal of Parallel and Distributed Computing
Variational cumulant expansions for intractable distributions

Journal of Artificial Intelligence Research
Topology selection for stream mining systems

LS-MMRM '09 Proceedings of the First ACM workshop on Large-scale multimedia retrieval and mining

Quantified Score

Hi-index	0.00

Visualization

Abstract

We introduce a large family of Boltzmann machines that can be trained by standard gradient descent. The networks can have one or more layers of hidden units, with tree-like connectivity. We show how to implement the supervised learning algorithm for these Boltzmann machines exactly, without resort to simulated or mean-field annealing. The stochastic averages that yield the gradients in weight space are computed by the technique of decimation. We present results on the problems of N-bit parity and the detection of hidden symmetries.