Class List

Here are the classes, structs, unions and interfaces with brief descriptions:

[detail level 123]

mic
algorithms
MazeHistogramFilter	Class implementing a histogram filter based solution of the maze-of-digits localization problem
application
EpisodicHistogramFilterMazeLocalization	Application for episodic testing of convergence of histogram filter based maze-of-digits localization
GridworldDeepQLearning	Class responsible for solving the gridworld problem with Q-learning and (not that) deep neural networks
GridworldDRLExperienceReplay	Class responsible for solving the gridworld problem with Q-learning, neural network used for approximation of the rewards and experience replay using for (batch) training of the neural network
GridworldDRLExperienceReplayPOMDP	Class responsible for solving the gridworld problem with Q-learning, neural network used for approximation of the rewards and experience replay using for (batch) training of the neural network. In this case there is an assumption that the agent observes only part of the environment (POMPD)
GridworldQLearning	Class responsible for solving the gridworld problem with Q-learning
GridworldValueIteration	Class responsible for solving the gridworld problem by applying the reinforcement learning value iteration method
HistogramFilterMazeLocalization	Class implementing a histogram filter based solution of the maze-of-digits problem
MazeOfDigitsDLRERPOMPD	Application of Partially Observable Deep Q-learning with Experience Reply to the maze of digits problem. There is an assumption that the agent observes only part of the environment (POMPD)
MNISTDigitDLRERPOMDP	Application of Partially Observable Deep Q-learning with Experience Reply to the MNIST digits problem. There is an assumption that the agent observes only part of the environment - a patch of the whole image (POMPD)
TestApp	Class implementing a n-Armed Bandits problem solving the n armed bandits problem using simple Q-learning rule
nArmedBanditsSofmax	Class implementing a n-Armed Bandits problem solving the n armed bandits problem using Softmax Action Selection
nArmedBanditsUnlimitedHistory	Class implementing a n-Armed Bandits problem solving the n armed bandits problem based on unlimited history action selection (storing all action-value pairs)
environments
Environment	Abstract class representing an environment
Gridworld	Class emulating the gridworld environment
MazeOfDigits	Class emulating the maze of digits environment
MNISTDigit	Class emulating the MNISTDigit digit environment
importers
MazeMatrixImporter
types
SpatialExperience	Structure storing a spatial experience - a triplet of position in time t, executed action and position in time t+1
SpatialExperienceMemory	Class representing the spatial experience memory - used in memory replay. Derived from the Batch class