ELeaRNT Evolutionary Learning of Rich Neural

Network Topologies

- Authors
- Slobodan Miletic 3078/2010 sloba10_at_gmail.com
- Nikola Jovanovic 3077/2010 nikolaj_ub_at_yahoo.co.u

k

Introduction

- Genetic algorithm
- mimics natural evolution
- candidate solution
- mutation
- Neural network
- based on biological neurons
- network consists of neurons grouped in layers

Problem definition

- Current design methods are manual and inefficient

- Hard to define number of neurons and connections
- No automated design method
- for specific optimal topology

Problem importance

- Neural networks have a large use area
- Creating new neural networks
- takes money and time

Problem trend

- Computers are getting more powerful
- New neural network uses are found
- If not solved, this problem would slow down the

evolution of neural networks

Existing solutions

- Trial Error
- Manual algorithm
- Few parameters for optimization
- Long, costly, and not very efficient

Existing solutions

- Destructive Algorithm
- Starts with very big networks
- Gets results by pruning the initial network
- A lot of time is spent on unnecessary training
- of big networks

Existing solutions

- Constructive Algorithm
- Starts with a small neural network
- Adds nodes and connections
- Uses input/error rate to form new nodes
- Can miss optimal solution

Proposed solution

- Whats better?
- General algorithm fitness function change

enables generation - of different neural networks types
- Created neural networks outperform
- neural network models designed by hand
- Besides the best network, it creates several

suboptimal networks - that can also be used as a solutions

Proposed solution

- Whats new?
- No similar general algorithm on the market
- Original set of genetic algorithm mutations

Proposed solution

- Whats its future?
- Computer power and parallelism are increasing
- which allows more complex neural networks
- Automated neural network generation algorithm
- like this will allow creation
- of complex neural networks

Solution details

3

1 point Crossover

2 point Crossover

5

g

3

f

Solution details

1 point Crossover

2 point Crossover

1 point Crossover

Solution details

5

g

b

2 point Crossover

1

a

Conclusion

- New way to create neural networks
- Results fully comparable
- with hand designed networks
- Space for further improvement
- More then one created network can be used

- QUESTIONS?