i am using ga to find the initial weight of ann, So i need know about the fitness function?

2 views (last 30 days)
i want to find the initial weight of ann by using genetic algorithm. i have multiple input to ann. how i link ann and ga to perform my task.

Accepted Answer

Matthew Eicholtz
Matthew Eicholtz on 12 May 2017
It sounds like you need to decide on a fitness function to evaluate candidate solutions provided by the genetic algorithm. Perhaps the easiest function is whatever you would use during training of the neural network (e.g., sum of the squared error on training data).
Here is just one of many potential solutions to this problem:
  1. Generate population of solutions for the genetic algorithm (GA).
  2. For each candidate solution, initialize and train a neural network with those corresponding weights.
  3. Test each trained neural network using some validation data.
  4. Compute some measure of error or accuracy (i.e., fitness) of the networks on the validation data.
  5. Use the fitness scores to rank the population.
  6. Apply GA operators (e.g., crossover, mutation, elitism) to create next generation of solutions.
  7. Repeat steps 2-7 until some stopping criterion is met.
Does this help?
  3 Comments
Greg Heath
Greg Heath on 15 May 2017
It is rare that you cannot find helpful posts in both the NEWSREADER and ANSWERS. A few search-word examples
greg neural genetic
greg neural ga
neural genetic
neural ga
Hope this helps.
Greg

Sign in to comment.

More Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!