2009 Semester Project
Traveling from the start to the goal in a gridworld which has a cliff
Sorry, I was not able to complete this document before the departure to Kaunas. Please start with reading Tutorials below.
To communicate, I want your e-mail address. Please everyone send me an e-mail to let me know your address, or someone collect email address of this group and send it to me.
In this Semester Project we study machine learning teqniques such as Genetic Algorithm,
Reinforcement Learning, Neural Network (McCulloh-Pits Neurons with Sigmoid or Spiking Neurons),
Ant Colony Optimization and so on, through so-called a "Cliff Walking Problem."
- Reference
Via Google, you may collect articles and/or web pages related to the topics you are interested in. For example with keywords "cliff-walk+gridworld+genetic-algorithm" The below finds such papers I collected and I think those are quite good.
- Reinforcement Learning (RL)
- Ant Colony Optimization (ACO)
The most populer application of ACO is to Traveling Saleseperson Problem (TSP). It would be good to start with learning ACO via TSP. See for example
TSP by ANT.
If you want to know ACO further in detail again long but good tutorial is here
Tutorial.
- Neural Networks (NN) with Sigmoid