CS 5970 Homework 6 — Neuroevolution for Problem Solving

Due Monday, October 19, 2016

1. Motivation

Just as with evolutionary computation in general, to use neuroevolution to solve problems, we need to apply them to particular problems. To do that, we need to take the same actions with regard the evolutionary computation aspects of neuroevolution as we do for any other application of evolutionary computation (following De Jong, Evolutionary Computation: A Unified Approach, p 72):

In addition, we need to make several decisions regarding what features of our artificial neural networks (ANNs) will be evolved, such as connection weights or various aspects of the structure of the network.

2. Goals

The goals of this assignment are:

3. Assignment

  1. List and explain at least 10 distinct features of an ANN that could be evolved using neuroevolution.
  2. Read the paper "Evolution of Robot-to-Robot Nurturing and Nurturability" by Leonce, Hoke, and Hougen and answer the questions below.

    1. What does an individual in the population represent?
    2. Is this a fixed-length linear object, a fixed-length nonlinear object, a variable-length linear object, or a nonlinear variable-length object?
    3. How is each individual encoded?
    4. Is this encoding genotypical or phenotypical?
    5. How is fitness calculated for each individual in the population?
    6. What reproductive operators are used?
    7. What parameters are used for each of the reproductive operators?
    8. What is the parent population size?
    9. What is the offspring population size?
    10. Is an overlapping or non-overlapping generation model used?
    11. What stopping criteria are used?
    12. What answers are returned?

4. What to Turn In

You will turn in to the appropriate dropbox in D2L a machine readable electronic copy of your homework that completes the exercises above.