+ All Categories
Home > Documents > The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Date post: 13-Dec-2015
Category:
Upload: cecilia-pierce
View: 213 times
Download: 0 times
Share this document with a friend
14
The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013
Transcript
Page 1: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

The Essence of PDP: Local Processing, Global Outcomes

PDP ClassJanuary 16, 2013

Page 2: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Goodness of Network States and their Probabilities

• Goodness of a network state• How networks maximize goodness• The Hopfield network and Rumelhart’s continuous

version• Stochastic networks: The Boltzmann Machine, and the

relationship between goodness and probability• Equilibrium, ergodicity, and annealing• Exploring the relationship between Goodness and

Probability in an ensemble of networks

Page 3: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Network Goodness and How to Increase it

Page 4: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

The Hopfield Network

• Assume symmetric weights.• Units have binary states [+1,-1]• Units are set into initial states• Choose a unit to update at random• If net > 0, then set state to 1.• Else set state to -1.• Goodness always increases… or stays the

same.

Page 5: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Rumelhart’s Continuous VersionUnit states have values between 0 and 1. Units are updated asynchronously. Update is gradual, according to the rule:

There are separate scaling parameters for external and internal input:

Page 6: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

The Cube Network

Positive weights have value +1Negative weights have value -1.5‘External input’ is implemented as a positive bias of .5 to all units.These values are all scaled by the istr parameter in calculating goodness in the program (istr = 0.4).

Page 7: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Goodness Landscape of Cube Network

Page 8: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Rumelhart’s Room Schema Model

• Units for attributes/objects found in rooms• Data: lists of attributes found in rooms• No room labels• Weights and biases:

• Modes of use in simulation:– Clamp one or more units, let the network settle– Clamp all units, let the network calculate the

Goodness of a state (‘pattern’ mode)

Page 9: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Weights for all units

Page 10: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Goodness Landscape for Some Rooms

Page 11: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Slices thru landscape with three different starting points

Page 12: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

The Boltzmann Machine:The Stochastic Hopfield Network

Units have binary states [0,1], Update is asynchronous. The activation function is:

Assuming processing is ergodic: that is, it is possible to get from any state to anyother state, then when the state of the network reaches equilibrium, the relative probability and relative goodness of two states are related as follows:

More generally, at equilibrium we have the Probability-Goodness Equation:

or

Page 13: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Simulated Annealing

• Start with high temperature. This means it is easy to jump from state to state.

• Gradually reduce temperature.• In the limit of infinitely slow annealing, we

can guarantee that the network will be in the best possible state (or in one of them, if two or more are equally good).

• Thus, the best possible interpretation can always be found (if you are patient)!

Page 14: The Essence of PDP: Local Processing, Global Outcomes PDP Class January 16, 2013.

Exploring Probability Distributions over States

• Imagine settling to a non-zero temperature, such as T = 0.5.

• At this temperature, there’s still some probability of being in a state that is less than perfect.

• Consider an ensemble of networks– At equilibrium (i.e. after enough cycles, possibly with annealing)

the relative frequencies of being in the different states will approximate the relative probabilities given by the Probability-Goodness equation.

• You will have an opportunity to explore this situation in the homework assignment


Recommended