Title: CAP6938 Neuroevolution and Artificial Embryogeny Basic Concepts
1CAP6938Neuroevolution and Artificial
EmbryogenyBasic Concepts
- Dr. Kenneth Stanley
- January 11, 2006
2We Care About Evolving ComplexitySo Why Neural
Networks?
- Historical origin of ideas in evolving complexity
- Representative of a broad class of structures
- Illustrative of general challenges
- Clear beneficiary of high complexity
3How Do NNs Work?
Output
Output
Input
Input
4How do NNs Work?Example
Outputs (effectors/controls)
Forward Left Right
Front Left Right Back
Inputs (Sensors)
5What Exactly Happens Inside the Network?
Neuron j activation
out1
out2
H1
H2
w11
w22
w21
w12
X1
X2
6Recurrent Connections
- Recurrent connections are backward connections in
the network - They allow feedback
- Recurrence is a type of memory
7Activating Networks of Arbitrary Topology
- Standard method makes no distinction between
feedforward and recurrent connections - The network is then usually activated once per
time tick - The number of activations per tick can be
- thought of as the speed of thought
- Thinking fast is expensive
out
Wout-H
wH-out
H
w21
w11
X1
X2
8Arbitrary Topology Activation Controversy
- The standard method is not necessarily the best
- It allows delay-line memory and a very simple
activation algorithm with no special case for
recurrence - However, all-at-once activation utilizes the
entire net in each tick with no extra cost - This issue is unsettled
9The Big Questions
- What is the topology that works?
- What are the weights that work?
?
?
?
?
?
?
?
?
?
?
?
?
?
10Problem Dimensionality
- Each connection (weight) in the network is a
dimension in a search space - The space youre in matters Optimization is not
the only issue! - Topology defines the space
21-dimensional space
3-dimensional space
11High Dimensional Space is Hard to Search
- 3 dimensional easy
- 100 dimensional need a good optimization method
- 10,000 dimensional very hard
- 1,000,000 dimensional very very hard
- 100,000,000,000,000 dim. forget it
12Bad News
- Most interesting solutions are high-D
- Robotic Maid
- World Champion Go Player
- Autonomous Automobile
- Human-level AI
- Great Composer
- We need to get into high-D space
13A Solution (preview)
- Complexification Instead of searching directly
in the space of the solution, start in a smaller,
related space, and build up to the solution - Complexification is inherent in vast examples of
social and biological progress
14So how do computers optimize those weights anyway?
- Depends on the type of problem
- Supervised Learn from input/output examples
- Reinforcement Learning Sparse feedback
- Self-Organization No teacher
- In general, the more feedback you get, the easier
the learning problem - Humans learn language without supervision
15Significant Weight Optimization Techniques
- Backpropagation Change weights based on their
contibution to error - Hebbian learning Changes weights based on firing
correlations between connected neurons
Homework -Fausett pp. 39-80 (in Chapter 2)-and
Fausett pp. 289-316 (in Chapter 6) -Online intro
chaper on RL -Optional RL survery