Laurent Itti: CS564 - Brain Theory and Artificial Intelligence

1
Laurent Itti CS564 - Brain Theory and
Artificial Intelligence

• Lecture 1. Introduction and Brain Overview
• Reading Assignments
• TMB2 Chapters 1 2.4
• HBTNN
• I.1. Introducing the Neuron (Arbib)
• Unless indicated otherwise, the TMB2 material
  is the required reading, and the other readings
  supplementary.

2
CS 564 Brain Theory and Artificial Intelligence

• URL http//iLab.usc.edu/classes/2002cs564/ for
  syllabus, instructor and TA information,
  handouts, homework and grades
• Instructor
• Laurent Itti itti_at_pollux (Office Hour Mon
  3-5, HNB30A)
• TA
• Yoo-Hee Shin yooheesh_at_usc.edu
• This course provides a basic understanding of
  brain function, and of artificial neural networks
  which provide tools for a new paradigm for
  adaptive parallel computation.
• No background in neuroscience is required, nor is
  specific programming expertise, but knowledge of
  C will be useful for homeworks.

3
Texts and Grading

• Text M.A. Arbib, 1989, The Metaphorical Brain
  2 Neural Networks and Beyond,
  Wiley-Interscience.
• Supplementary reading
• M.A. Arbib, Ed., 1995, The Handbook of Brain
  Theory and Neural Networks, MIT Press
  (paperback).
• One mid-term and a final will cover the entire
  contents of the readings so far as well as the
  lectures.
• The final exam will cover all of the course, but
  emphasizing material not covered in the mid-term.
• Distribution of Grades
• Homeworks 40 Mid-term 30 Final Exam 30.

4
Syllabus Overview

• Introduction
• Overview
• Charting the brain
• The Brain as a Network of Neurons

5
Syllabus Overview

• Introduction (cont.)
• Experimental techniques
• Introduction to Vision
• Schemas

6
Syllabus Overview

• Basic Neural Modeling Adaptive Networks
• Didday Model of Winner-Take-All
• Hopfield networks
• Adaptive networks Hebbian learning
• Perceptrons landmark learning

E - ½ ? ij sisjwij ? i siqi
7
Syllabus Overview

• Neural Modeling Adaptive Networks (cont.)
• Adaptive networks gradient descent
• and backpropagation
• Reinforcement learning
• Competition and cooperation
• Visual plasticity self-organizing
• feature maps Kohonen maps

? ? ? ? ? ? ? ? ? ? ? ?
? ? ? ? ? ?
8
Syllabus Overview

• Examples of Large-scale Neural Modeling
• System concepts
• Model of saccadic eye movements
• Feedback and the spinal cord
• mass-spring model of muscle

9
Syllabus Overview

• Large-scale Neural Models of Vision
• Early visual processing
• Depth perception
• Optic flow

10
Syllabus Overview

• Large-scale Neural Models of Vision (cont.)
• Visual attention
• Object recognition
• Scene perception

11
Syllabus Overview

• Other Advanced Neural Modeling
• Reaching, grasping and affordances
• Cerebellar adaptation
• Memory and consciousness

12
Syllabus Overview

• Applications and Outlook
• Towards highly-capable
• robots
• Overview and summary

13
Three Frameworks

• Artificial intelligence (AI) build a packet of
  intelligence into a machine
• Cognitive psychology explain human behavior by
  interacting processes (schemas) in the head but
  not localized in the brain
• Brain Theory interactions of components of the
  brain -
• - computational neuroscience
• - neurologically constrained models e.g.,
  networks of neurologically localized schemas
• and abstracting from them as both Artificial
  intelligence and Cognitive psychology
• - connectionism networks of trainable
  quasi-neurons to provide parallel distributed
  models little constrained by neurophysiology
• - abstract (computer program or control system)
  information processing models

14
The Aim of the Course

• To gain an understanding of biological neurons as
  the basis for
• Brain Theory modeling interactions of components
  of the brain, especially more or less realistic
  biological neural networks localized in specific
  brain regions
• Connectionism in both Artificial intelligence
  (AI) and Cognitive psychology modeling
  artificial neural networks -- networks of
  trainable quasi-neurons -- to provide parallel
  distributed models of intelligence in humans,
  animals and machines
• This lecture A tourists guide to the brain -)

15
A motivating theme Vision

• Vision as a progressive change in representation
• Marr (1982) through 2 ½ D primal sketch
• Because vision is by far the most studied sense
  (because it is easy to experiment with), we will
  use it as a basis for many examples of models
  studied in this course.

16
Vision and the brain
Ryback et al, 1998

• Roughly speaking, about half of
• the brain is concerned with vision.
• Although most of it is highly auto-
• mated and unconscious, vision hence
• is a major component of brain function.

17
Vision, AI and robots

• 1940s beginning of Artificial Intelligence
• McCullogh Pitts, 1942
• Si wixi ? q
• Perceptron learning rule (Rosenblatt, 1962)
• Backpropagation
• Hopfield networks (1982)
• Kohonen self-organizing maps

18
Vision, AI and Robots

• 1950s beginning of computer vision
• Aim give to machines same or better vision
  capability as ours
• Drive AI, robotics applications and factory
  automation
• Initially passive, feedforward, layered and
  hierarchical process
• that was just going to provide input to higher
  reasoning
• processes (from AI)
• But soon realized that could not handle real
  images
• 1980s Active vision make the system more robust
  by allowing the
• vision to adapt with the ongoing
  recognition/interpretation

19
A tourists guide to the brain

• Gross anatomy
• Non-neural structures
• Major cortical areas

20
Central vs. Peripheral Nervous System

• The brain is not the entire nervous systems
  there is also
• the spinal cord, many peripheral ganglia (small
  
• clusters of neurons), and neurons extend
  connections to
• locations all over the body (e.g., sensory
  neurons, motor neurons).

21
Autonomic Nervous System
22
(No Transcript)
23
Axes in the brain
24
The Bauplan for the Mammalian Brain
25
Medical Orientation Terms for Slices
26
Main Arterial Supply to the Brain
27
Arterial Supply is Segmented

• Occlusion/damage to one artery will affect
  specific brain
• regions. Important to remember for patient
  studies.

28
Ventricular System

• Ventricules Cavities filled with fluid inside
  and around
• the brain. One of their functions is to drain
  garbage out
• of the brain.

29
Cortical Lobes
Sulcus (fissure if very large) Grooves in
folded cortex Gyrus cortex between two sulci 1
sulcus, many sulci 1 gyrus, many gyri
30
(No Transcript)
31
(No Transcript)
32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
Neurons

• Cell body (soma) where computation takes place
• Dendrites input branches
• Axon unique output (but may branch out)
• Synapse connection between presynaptic axon and
• postsynaptic dendrite (in general).

37
Electron Micrograph of a Real Neuron
38
Neurons and Synapses
39
Grey and White Matters

• Grey matter neurons (cell bodies), at outer
  surface of brain
• White matter interconnections, inside the brain
• Deep nuclei clusters of neurons deep inside the
  brain

40
Major Functional Areas

• Primary motor voluntary movement
• Primary somatosensory tactile, pain, pressure,
  position, temp., mvt.
• Motor association coordination of complex
  movements
• Sensory association processing of multisensorial
  information
• Prefrontal planning, emotion, judgement
• Speech center (Brocas area) speech production
  and articulation
• Wernickes area comprehen-
• sion of speech
• Auditory hearing
• Auditory association complex
• auditory processing
• Visual low-level vision
• Visual association higher-level
• vision

41
Major Functional Areas
42
A View of the Monkey Brain
43
(No Transcript)
44
http//www.radiology.wisc.edu/Med_Students/neurora
diology/fmri/
45
(No Transcript)
46
(No Transcript)
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
Limbic System

• Cortex inside the brain.
• Involved in emotions, sexual behavior, memory,
  etc
• (not very well known)

53
Major Functional Areas
54
Visual Input to the Brain
55
Eye and retina
56
Human Visual System
57
Primary Visual Pathway
58
Layered Organization of Cortex

• Cortex is 1 to 5mm-thick, folded at the surface
  of the brain
• (grey matter), and organized as 6 superimposed
  layers.
• Layer names
• 1 Molecular layer
• 2 External granular layer
• 3 External pyramidal layer
• 4 internal granular layer
• 5 Internal pyramidal layer
• 6 Fusiform layer
• Basic layer functions
• Layers 1/2 connectivity
• Layer 4 Input
• Layers 3/5 Pyramidal cell bodies
• Layers 5/6 Output

59
Layered Organization of Cortex
60
Slice through the thickness of cortex
1 2 3 4 5 6
61
Columnar Organization

• Very general principle in cortex neurons
  processing similar things are grouped together
  in small patches, or columns, or cortex.
• In primary visual cortex
  as in higher (object recognition)
  visual areas
• and in many, non-visual, areas as well (e.g.,
  auditory, motor, sensory, etc).

62
Retinotopy

• Many visual areas are organized as retinotopic
  maps locations next
• to each other in the outside world are
  represented by neurons close
• to each other in cortex.
• Although the topology is thus preserved, the
  mapping typically is highly non-linear (yielding
  large deformations in representation).
• Stimulus shown on screen
  and corresponding activity in cortex!

63
Retinotopy
64
Mammalian and Frog Visual Systems
65
(No Transcript)
66
(No Transcript)
67
Interconnect
Felleman Van Essen, 1991
68
Interconnect (other source)
69
More on Connectivity
70
Varieties of Vertebrate Brains
Catfish
Snake
Alligator
Frog
Primitive Mammal
Goose
Horse
71
Outlook

• There is a lot to learn about the brain!
• but dont feel overwhelmed, we will smoothly
• introduce all new concepts.
• Principled theoretical and engineering methods
  will allow us to abstract some of these
  complications.
• Starting with fundamental techniques, we will
  then study fairly complex, large-scale neural
  models.