Design of self organizing learning array for intelligent machines
This presentation is the property of its rightful owner.
Sponsored Links
1 / 48

Design of Self-Organizing Learning Array for Intelligent Machines PowerPoint PPT Presentation


  • 101 Views
  • Uploaded on
  • Presentation posted in: General

Design of Self-Organizing Learning Array for Intelligent Machines. Janusz Starzyk School of Electrical Engineering and Computer Science Heidi Meeting June 3 2005. Motivation: How a new understanding of the brain will lead to the creation of truly intelligent machines

Download Presentation

Design of Self-Organizing Learning Array for Intelligent Machines

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


Design of self organizing learning array for intelligent machines

Design of Self-Organizing Learning Array for Intelligent Machines

Janusz Starzyk

School of Electrical Engineering and Computer Science

Heidi Meeting June 3 2005

Motivation:

How a new understanding of the brain will lead

to the creation of truly intelligent machines

fromJ. Hawkins “On Intelligence”


Design of self organizing learning array for intelligent machines

Elements of Intelligence

  • Abstract thinking and action planning

  • Capacity to learn and memorize useful things

  • Spatio-temporal memories

  • Ability to talk and communicate

  • Intuition and creativity

  • Consciousness

  • Emotions and understanding others

  • Surviving in complex environment and adaptation

  • Perception

  • Motor skills in relation to sensing and anticipation


Problems of classical ai

Problems of Classical AI

  • Lack of robustness and generalization

  • No real-time processing

  • Central processing of information by a single processor

  • No natural interface to environment

  • No self-organization

  • Need to write software


Intelligent behavior

Intelligent Behavior

  • Emergent from interaction with environment

  • Based on large number of sparsely connected neurons

  • Asynchronous

  • Self-timed

  • Interact with environment through sensory-motor system

  • Value driven

  • Adaptive


Design principles of intelligent systems

Design principles of intelligent systems

from Rolf Pfeifer “Understanding of Intelligence”

Design principles

  • synthetic methodology

  • time perspectives

  • emergence

  • diversity/compliance

  • frame-of-reference

Agent design

complete agent principle

cheap design

ecological balance

redundancy principle

parallel, loosely coupled processes

sensory-motor coordination

value principle


The principle of cheap design

The principle of “cheap design”

  • intelligent agents: “cheap”

    • exploitation of ecological niche

    • economical (but redundant)

    • exploitation of specific physical properties of interaction with real world


Principle of ecological balance

Principle of “ecological balance”

  • balance / task distribution between

    • morphology

    • neuronal processing (nervous system)

    • materials

    • environment

  • balance in complexity

    • given task environment

    • match in complexity of sensory, motor, and neural system


The redundancy principle

The redundancy principle

  • redundancy prerequisite for adaptive behavior

  • partial overlap of functionality in different subsystems

  • sensory systems: different physical processes with “information overlap”


Generation of sensory stimulation through interaction with environment

Generation of sensory stimulation through interaction with environment

  • multiple modalities

  • constraints from morphology and materials

  • generation of correlations through physical process

  • basis for cross-modal associations


The principle of sensory motor coordination

The principle of sensory-motor coordination

  • Holk Cruse

    • •no central control

    • •only local neuronal communication

    • •global communication through environment

    • neuronal connections

  • self-structuring of sensory data through interaction with environment

  • physical process —not „computational“

  • prerequisite for learning


The principle of parallel loosely coupled processes

The principle of parallel, loosely coupled processes

  • Intelligent behavior emergent from agent-environment interaction

  • Large number of parallel, loosely coupled processes

  • Asynchronous

  • Coordinated through agent’s

    –sensory-motor system

    –neural system

    –interaction with environment


The value principle

The “value principle”

  • about motivation

  • evaluation of actions

  • frame-of-reference: explicit and implicit values

  • recent theorizing: information theoretic

  • (organism tries to mainting “flow of information”)


Design of self organizing learning array for intelligent machines

Human Brain at Birth

14 Years Old

6 Years Old

Neuron Structure and Self-Organizing Principles

13


Neuron structure and self organizing principles cont d

Neuron Structure and Self-Organizing Principles (Cont’d)


Design of self organizing learning array for intelligent machines

Motor cortex

Somatosensory cortex

Sensory associative

cortex

Pars

opercularis

Visual associative

cortex

Broca’s

area

Visual

cortex

Primary

Auditory cortex

Wernicke’s

area

Brain Organization


Minicolumn organization and self organizing learning arrays

Minicolumn Organization and Self Organizing Learning Arrays

  • V. Mountcastle argues that all regions of the brain perform the same algorithm

  • SOLAR combines many groups of neurons (minicolumns) in a pseudorandom way

  • Each microcolumn has the same structure

  • Thus it performs the same computational algorithm satisfying Mountcastle’s principle

  • VB Mountcastle (2003). Introduction [to a special issue of Cerebral Cortex on columns]. Cerebral Cortex, 13, 2-4.


Design of self organizing learning array for intelligent machines

Cortical Minicolumns

“The basic unit of cortical operation is the minicolumn… It contains of the order of 80-100 neurons except in the primate striate cortex, where the number is more than doubled. The minicolumn measures of the order of 40-50 m in transverse diameter, separated from adjacent minicolumns by vertical, cell-sparse zones … The minicolumn is produced by the iterative division of a small number of progenitor cells in the neuroepithelium.” (Mountcastle, p. 2)

Stain of cortex in planum temporale.


Design of self organizing learning array for intelligent machines

Groupping of Minicolumns

Groupings of minicolumns seem to form the physiologically observed functional columns. Best known example is orientation columns in V1.

They are significantly bigger than minicolumns, typically around 0.3-0.5 mm and have 4000-8000 neurons

Mountcastle’s summation:

“Cortical columns are formed by the binding together of many minicolumns by common input and short range horizontal connections. … The number of minicolumns per column varies … between 50 and 80. Long range intracortical projections link columns with similar functional properties.” (p. 3)


Sparse connectivity

Sparse Connectivity

The brain is sparsely connected.

(Unlike most neural nets.)

A neuron in cortex may have on the order of 100,000 synapses. There are more than 1010 neurons in the brain. Fractional connectivity is very low: 0.001%.

Implications: 

  • Connections are expensive biologically since they take up space, use energy, and are hard to wire up correctly.

  • Therefore, connections are valuable.

  • The pattern of connection is under tight control.

  • Short local connections are cheaper than long ones.

    Our approximation makes extensive use of local connections for computation.


Design of self organizing learning array for intelligent machines

Introducing Self-Organizing Learning Array SOLAR

  • SOLAR is a regular array of identical processing cells, connected to programmable routing channels.

  • Each cell in the array has ability to self-organize by adapting its functionality in response to information contained in its input signals.

  • Cells choose their input signals from the adjacent routing channels and send their output signals to the routing channels.

  • Processing cells can be structured to implement minicolumns


Solar hardware architecture

SOLAR Hardware Architecture


Solar routing scheme

SOLARRouting Scheme


Pcb solar

PCB SOLAR

XILINX

VIRTEX XCV 1000


Design of self organizing learning array for intelligent machines

System SOLAR


Wiring in solar

Wiring in SOLAR

Initial wiring and final wiring selection for credit card

approval problem


Solar classification results

SOLAR Classification Results


Design of self organizing learning array for intelligent machines

Associative SOLAR


Design of self organizing learning array for intelligent machines

Associations made in SOLAR


Defining simple brain

Sensors

Actuators

Defining Simple Brain

Reactive

Associations

Sensory

Inputs

Motor

Outputs


Simple brain properties

Simple Brain Properties

  • Interacts with environment through sensors and actuators

  • Uses distributed processing in sparsely connected neurons

  • Uses spatio-temporal associative learning

  • Uses feedback for input prediction and screening input information for novelty


Brain structure with value system

Sensors

Actuators

Brain Structure with Value System

Value System

Action

Planning

Reinforcement

Signal

Anticipated Response

Sensory

Inputs

Motor

Outputs


Brain structure with value system properties

Brain Structure with Value System Properties

  • Interacts with environment through sensors and actuators

  • Uses distributed processing in sparsely connected neurons

  • Uses spatio-temporal associative learning

  • Uses feedback for input prediction and screening input information for novelty

  • Develops an internal value system to evaluate its state in environment using reinforcement learning

  • Plans output actions for each input to maximize the internal state value in relation to environment

  • Uses redundant structures of sparsely connected processing elements


Value system in reinforcement learning control

Value System in Reinforcement Learning Control

States

Controller

Environment

Value System

Optimization

Reinforcement Signal


Artificial brain organization

Value System

Action

Planning

Sensors

Actuators

Artificial Brain Organization

Understanding

Decision making

Anticipated Response

Reinf.

Signal

Motor

Outputs

Sensory

Inputs


Artificial brain organization1

Artificial Brain Organization

  • Learning should be restricted to unexpected situation or reward

  • Anticipated response should have expected value

  • Novelty detection should also apply to the value system

  • Need mechanism to improve and compare the value


Artificial brain organization2

Value System

Action

Planning

Sensors

Actuators

Artificial Brain Organization

Understanding

Improvement

Detection

Expectation

Comparison

Inhibition

Novelty

Detection

Anticipated Response

Reinf.

Signal

Motor

Outputs

Sensory

Inputs


Artificial brain organization3

Artificial Brain Organization

  • Anticipated response block should learn the response that improves the value

  • A RL optimization mechanism may be used to learn the optimum response for a given value system and sensory input

  • Random perturbation should be applied to the optimum response to explore possible states and learn their the value

  • New situation will result in new value and WTA will chose the winner


Artificial brain organization4

Artificial Brain Organization

Positive

Reinforcement

Negative

Reinforcement

Sensory

Inputs

Motor

Outputs


Artificial brain selective processing

Artificial Brain Selective Processing

  • Sensory inputs are represented by more and more abstract features in the sensory inputs hierarchy

  • Possible implementation is to use winner takes all or Hebbian circuits to select the best match

  • Random wiring may be used to preselect sensory features

  • Uses feedback for input prediction and screening input information for novelty

  • Uses redundant structures of sparsely connected processing elements


Microcolumn organization

Microcolumn Organization

superneuron

WTA

Positive

Reinforcement

Negative

Reinforcement

WTA

WTA

Sensory

Inputs

Motor

Outputs


Design of self organizing learning array for intelligent machines

Superneuron Organization

  • Each microcolumn contains a number of superneurons

  • Within each microcolumn, superneurons compete on different levels of signal propagation

  • Superneuron contains a predetermined configuration of

    • Sensory (blue)

    • Motor and (yellow)

    • Reinforcement neurons (positive green and negative red)

  • Superneurons internally organize to perform operations of

    • Input selection and recognition

    • Association of sensory inputs

    • Feedback based anticipation

    • Learning inhibition

    • Associative value learning, and

    • Value based motor activation


Design of self organizing learning array for intelligent machines

Superneuron Organization

  • Sensory neurons are primarily responsible for providing information about environment

    • They receive inputs from sensors or other sensory neurons on lower level

    • They interact with motor neurons to represent action and state of environment

    • They provide an input to reinforcement neurons

    • They help to activate motor neurons

  • Motor neurons are primarily responsible for activation of motor functions

    • They are activated by reinforcement neurons with the help from sensory neurons

    • They activate actuators or provide an input to lower level motor neurons

    • They provide an input to sensory neurons

  • Reinforcement neurons are primarily responsible for building the internal value system

    • They receive inputs from reinforcement learning sensors or other reinforcement neurons on lower level

    • They receive inputs from sensory neurons

    • They provide an input to motor neurons

    • They help to activate sensory neurons


Design of self organizing learning array for intelligent machines

S1h

S2h

WTA

S2

S3

WTA

WTA

S1

Sensory Neurons Interactions


Design of self organizing learning array for intelligent machines

WTA

WTA

WTA

Sensory Neurons Functions

  • Sensory neurons are responsible for

    • Representation of inputs from environment

    • Interactions with motor functions

    • Anticipation of inputs and screening for novelty

    • Selection of useful information

    • Identifying invariances

    • Making spatio-temporal associations


Design of self organizing learning array for intelligent machines

Sensory Neurons Functions

Sensory neurons

  • Represent inputs from environment by

    • Responding to activation from lower level (summation)

    • Selecting most likely scenario (WTA)

  • Interact with motor functions by

    • Responding to activation from motor outputs (summation)

  • Anticipate inputs and screen for novelty by

    • Correlation to sensory inputs from higher level

    • Inhibition of outputs to higher level

  • Select useful information by

    • Correlating its outputs with reinforcement neurons

  • Identify invariances by

    • Making spatio-temporal associations betweenneighbor sensory neurons


From apparent mess

From Apparent Mess


To clear mind organization

To Clear Mind Organization

WTA

WTA

WTA

WTA


  • Login