Deep Neural Networks - Notes for Lesson 11

Hopfield Nets and Boltzmann machines

This module deals with Boltzmann machine learning
deep learning
neural networks
notes
RBM
restricted Boltzmann machine
coursera
Author

Oren Bochman

Published

Saturday, October 21, 2017

{{< pdf lec11.pdf width="100%" class="ppSlide" >}}

Lecture 11a: Hopfield Nets

(Hopfield 1982)

Now, we leave behind the feedforward deterministic networks that are trained with backpropagation gradients. We’re going to see quite a variety of different neural networks now. These networks do not have output units. These networks have units that can only be in states 0 and 1. These networks do not have units of which the state is simply a function of the state of other units. These networks are, instead, governed by an “energy function”. Best way to really understand Hopfield networks: Go through the example of the Hopfield network finding a low energy state, by yourself. Better yet, think of different weights, and do the exercise with those. Typically, we’ll use Hopfield networks where the units have state 0 or 1; not -1 or 1.

Lecture 11b: Dealing with spurious minima

The last in-video question is not easy. Try to understand how the perceptron learning procedure is used in a Hopfield net; it’s not very thoroughly explained.

Lecture 11c: Hopfield nets with hidden units

This video introduces some sophisticated concepts, and is not entirely easy. An “excitatory connection” is a connection of which the weight is positive. “inhibitory”, likewise, means a negative weight. We look for an energy minimum, “given the state of the visible units”. That means that we look for a low energy configuration, and we’ll consider only configurations in which the visible units are in the state that’s specified by the data. So we’re only going to consider flipping the states of the hidden units. Be sure to really understand the last two sentences that Geoffrey speaks in this video.

Lecture 11e: How a Boltzmann machine models data

Now, we’re making a generative model of binary vectors. In contrast, mixtures of Gaussians are a generative model of real-valued vectors. 4:38: Try to understand how a mixture of Gaussians is also a causal generative model. 4:58: A Boltzmann Machine is an energy-based generative model. 5:50: Notice how this is the same as the earlier definition of energy. What’s new is that it’s mentioning visible and hidden units separately, instead of treating all units the same way.

References

Hopfield, John J. 1982. “Neural Networks and Physical Systems with Emergent Collective Computational Abilities.” Proceedings of the National Academy of Sciences 79 (8): 2554–58.

Reuse

CC SA BY-NC-ND

Citation

BibTeX citation:
@online{bochman2017,
  author = {Bochman, Oren},
  title = {Deep {Neural} {Networks} - {Notes} for {Lesson} 11},
  date = {2017-10-21},
  url = {https://orenbochman.github.io/notes/dnn/dnn-11/l_11.html},
  langid = {en}
}
For attribution, please cite this work as:
Bochman, Oren. 2017. “Deep Neural Networks - Notes for Lesson 11.” October 21, 2017. https://orenbochman.github.io/notes/dnn/dnn-11/l_11.html.