Man pages for markdumke/reinforcelearn
Reinforcement Learning

CliffWalkingCliff Walking
EligibilityEligibility traces
EnvironmentCustom Reinforcement Learning Environment
EpsilonGreedyPolicyEpsilon Greedy Policy
getEligibilityTracesGet eligibility traces
getReplayMemoryGet replay memory.
getStateValuesGet state values.
getValueFunctionGet weights of value function.
gridworldGridworld
GymEnvironmentGym Environment
interactInteraction between agent and environment.
makeAgentCreate Agent.
makeAlgorithmMake reinforcement learning algorithm.
makeEnvironmentCreate reinforcement learning environment.
makePolicyCreate policy.
makeReplayMemoryExperience Replay
makeValueFunctionValue Function Representation
MdpEnvironmentMDP Environment
MountainCarMountain Car
nHotMake n hot vector.
QLearningQ-Learning
RandomPolicyRandom Policy
reinforcelearnReinforcement Learning.
SoftmaxPolicySoftmax Policy
tilecodingTile Coding
ValueNetworkValue Network
ValueTableValue Table
windyGridworldWindy Gridworld
markdumke/reinforcelearn documentation built on Nov. 17, 2022, 12:53 a.m.