reinforcement-learning

I have neural network in tensorflow with trained parameters, it is "policy" for agent. Network is being updated in training loop in main tensorflow session in core program. In the end of each training cycle I need to pass this network to few parallel processes ("workers"), which will use it for collecting samples from interactions of agent's policy with environment. I need to do it in parallel, because simulating environment takes most of the time and runs only single-core. So, few parallel sampling processes are needed. I am struggling how to structure this in distributed tensorflow. What I

I'm writing an assisted learning algorithm in Java. I've run into a mathematical problem that I can probably solve, but because the processing will be heavy I need an optimum solution. That being said, if anyone knows a optimized library that will be totally awesome, but the language is Java so that will need to be taken into consideration. The idea is fairly simple: Objects will store combination of variables such as ABDC, ACDE, DE, AE. The max number of combination will be based on how many I can run without slowing down the program, so theoretically 100 lets say. The decision process will

I have a network in Keras with many outputs, however, my training data only provides information for a single output at a time. At the moment my method for training has been to run a prediction on the input in question, change the value of the particular output that I am training and then doing a single batch update. If I'm right this is the same as setting the loss for all outputs to zero except the one that I'm trying to train. Is there a better way? I've tried class weights where I set a zero weight for all but the output I'm training but it doesn't give me the results I expect? I'm using

How is Q-learning different from value iteration in reinforcement learning? I know Q-learning is model-free and training samples are transitions (s, a, s', r) . But since we know the transitions and the reward for every transition in Q-learning, is it not the same as model-based learning where we know the reward for a state and action pair, and the transitions for every action from a state (be it stochastic or deterministic)? I do not understand the difference. You are 100% right that if we knew the transition probabilities and reward for every transition in Q-learning, it would be pretty

I've seen such words as: A policy defines the learning agent's way of behaving at a given time. Roughly speaking, a policy is a mapping from perceived states of the environment to actions to be taken when in those states. But still didn't fully understand. What exactly is a policy in reinforcement learning? The definition is correct, though not instantly obvious if you see it for the first time. Let me put it this way: a policy is an agent's strategy . For example, imagine a world where a robot moves across the room and the task is to get to the target point (x, y), where it gets a reward.