Create a reinforcement learning agent for playing checkers game

Hi, I am Sourabh Jain, I am an experienced python programmer. I have also written feed forward neural network (fully connected) in python for analysis of stock price movements. For reinforcement learning, I will take target value and present value as function of sum of black/red pieces with weights . I will minimize the RMS error from target value and every move will update the weights using normalized error. If you would like to discuss this further, I am available. Regards Sourabh

Hi, I have just completed a project implementing multiplayer Checkers game in Python, and so can easily adapt the relevant code to this project if necessary. At university I studied artificial intelligence as part of my degree, the main assessment of the course being to implement a neural network, so I have experience in the sort of machine learning required as well as a theoretical background in other techniques which could be applied. Please get in touch if you wish to discuss the project in more detail.

Hi there! More information is needed before I can make a proper proposal. In particular I would like to know: 1. How comprehensive does the game need to be? Does it need to be displayed? If so, does it need to have a graphical display or would an ASCII display be fine? Will it be required for a human to play against the A.I.? 2. Is there any restriction/preference as to the type of machine learning algorithm? My preference is to use a Random Forest (using sklearn) off the shelf then to try out different types of Neural Networks (using theano). 3. What is the end-goal for this project? What needs to be handed-in or presented? The code? A playable A.I.? I have lots of other questions on top of these but I will wait to hear back before asking. Despite the lack of info the project sounds really cool and I would love to work on it :) I'm currently working on a human-in-the-loop-strategy to train a probability distribution classifier at work. This is all in python also and it sounds like there's a big overlap with this project. Hope to hear back from you, Ben

I really like this kind of work (using machine learning) . I have done a similar project in the past (checkers game using reinforcement learning). I'm sure I can deliver the work you need in the specified time.