Title: A Hierarchical Selforganizing Associative Memory for Machine Learning Janusz A' Starzyk, Ohio Univer
1A Hierarchical Self-organizing Associative Memory
forMachine LearningJanusz A. Starzyk, Ohio
UniversityHaibo He, Stevens Institute of
TechnologyYue Li, O2 Micro Inc
2Outline
- Introduction
- Associative learning algorithm
- Memory network architecture and operation
- Simulation analysis
- Conclusion and future research
3Introduction A biological point of view
Source The computational brain by P. S.
Churchland and T. J. Sejnowski
Memory is a critical component for understanding
and developing natural intelligent
machines/systems The question is How???
4Introduction self-organizing learning
array(SOLAR)
- Characteristics
- Self-organization
- Sparse and local interconnections
- Dynamically reconfigurable
- Online data-driven learning
5Introduction from SOLAR to AM
Feed forward Feed backward
Feed forward only
- Characteristics
- Self-organization
- Sparse and local interconnections
- Feedback propagation
- Information inference
- Hierarchical organization
- Robust and self-adaptive
- Capable of both hetero-associative (HA) and
auto-associative (AA)
6Outline
- Introduction
- Associative learning algorithm
- Memory network architecture and operation
- Simulation analysis
- Conclusion and future research
7Basic learning element
Self-determination of the function value
An example
8Signal strength (SS)
- Provides a coherent way to determine when to
trigger an association - Helps to resolve multiple feedback signals
Signal strength (SS) Signal value logic
threshold (SS range 0, 1)
9Three types of associations
- IOA Input only association
- OOA Output only association
- INOUA Input-output association
10Probability based associative learning algorithm
- Case 1
- Given the values of both inputs, decide the
output value
11Probability based associative learning algorithm
- Case 2
- Given the values of one input and an un-defined
output, decide the value of the other input
For instance
12Probability based associative learning algorithm
- Case 3
- Given the values of the output, decide the values
of both inputs
13Probability based associative learning algorithm
- Case 4
- Given the values of one input and the output,
decide the other input value
For instance
14Outline
- Introduction
- Associative learning algorithm
- Memory network architecture and operation
- Simulation analysis
- Conclusion and future research
15Network operations
Input data
Depth
Feedback operation
Feed forward operation
16Memory operation
Defined signal
Input data
1
3
Recovered signal
Signal resolved based on SS
5
4
2
Undefined signal
17Outline
- Introduction
- Associative learning algorithm
- Memory network architecture and operation
- Simulation analysis
- Conclusion and future research
18Hetero-associative memory Iris database
classification
3 classes, 4 numeric attributes, 150 instances
N-bits sliding-bar coding mechanism
Features
Class identity labels
In our simulation N80, L20, M30
19Neuron association pathway
Classification accuracy 96
20Auto-associative memory Panda image recovery
64 x 64 binary panda image
for a black pixel
for a white pixel
30 missing pixels
Error 0.4394
Error 2.42
Block half
Original image 64x64 binary image
21Outline
- Introduction
- Associative learning algorithm
- Memory network architecture and operation
- Simulation analysis
- Conclusion and future research
22Conclusion and future research
- Hierarchical associative memory architecture
- Probabilistic information processing,
transmission, association and prediction - Self-organization
- Self-adaptive
- Robustness
23Future research
Its all about design natural intelligent
machines !
- Multiple-inputs (gt2) association mechanism
- Dynamically self-reconfigurable
- Hardware implementation
- Facilitate goal-driven learning
- Spatio-temporal memory organization
How far are we???
3DANN
Brain On Silicon will not just be a dream or
scientific fiction in the future!
Picture source http//www.cs.utexas.edu/users/ai
-lab/fai/ and Irvine Sensors Corporation (Costa
Mesa, CA)