* @brief This class represents the double QLearning algorithm. * * The QLearning algorithm is biased to overestimate the expected future

* @brief This class represents the QLearning algorithm. * * This algorithm is a very simple but powerful way to learn the * optimal QFu

* @brief This class represents the SARSA algorithm. * * This algorithm is a very simple but powerful way to learn a * QFunction for an

* @brief This class represents the SARSAL algorithm. * * This algorithms adds eligibility traces to the SARSA algorithm. * * \sa S

* @brief This class models a cooperative MDP. * * This class can be used in order to model problems where multiple agents * cooperate i

* @brief This class solves an MDP using Linear Programming. * * This class is a very simple wrapper for solving an MDP using linear * p

* @brief This class implements the Linear Reward Penalty algorithm. * * This algorithm performs direct policy updates depending on whether a

* @brief The model for rPOMCP tests * * This model has been specially designed to require different answers * depending on whether rPOMCP is runnin

* @brief This class implements the Augmented MDP algorithm. * * This algorithm transforms a POMDP into an approximately equivalent * MD

* @brief This class implements the ADAM gradient descent algorithm. * * This class keeps things simple and fast. It takes two pointers to tw

* @brief This struct represents a basis function. * * A basis function is simply a function defined on some subset of factors. * It map

* @brief This struct represents a basis matrix. * * Note that the term "basis matrix" does not really exist in the * literature, it's j

* @brief This class implements the blind strategies lower bound. * * This class is useful in order to obtain a very simple lower bound for a

* @brief This class is used as the priority queue for CooperativePrioritizedSweeping. * * This class performs a similar work as that done by

* @brief This class can parse files containing MDPs and POMDPs in the Cassandra file format. */

* @brief This class keeps track of registered events and rewards. * * This class is a simple logger of events. It keeps track of both *

* @brief This class models CooperativeExperience as a CooperativeModel using Maximum Likelihood. * * Often an MDP is not known in advance. I

* @brief This class represents the Cooperative QLearning algorithm. * * This is the same as SparseCooperativeQLearning, but we handle dense

* @brief This class models CooperativeExperience as a CooperativeModel using Thompson Sampling. * * Often an MDP is not known in advance. It

* @brief The possible actions in a GridWorld-like environment. */

* @brief This class represents a Dynamic Decision Network with factored actions. * * This class is able to represent a Dynamic Decision Netw

* @brief This class represents the structure of a dynamic decision network. * * A DDN is a graph that relates how state features and agents

* @brief This class implements the Exploring Selfish Reinforcement Learning algorithm. * * This is a learning algorithm for common interest

* @brief This class represents an epsilon-greedy policy for Factored MDPs */

* @brief This class represents the ExpectedSARSA algorithm. * * This algorithm is a subtle improvement over the SARSA algorithm. *

* @brief This class computes averages and counts for a multi-agent cooperative Bandit problem. * * This class can be used to compute the ave

* @brief This class keeps track of registered events and rewards. * * This class is a simple aggregator of events. It keeps track of both th

Forward references to avoid including tons of headers

* @brief This class computes averages and counts for a Bandit problem. * * This class can be used to compute the averages and counts for all

* @brief This class represents the Factored LP algorithm. * * This algorithm has been introduced in a number of Guestrin et al. * paper

* @brief This class represents a factored 2D matrix. * * Note that we can't use a multi_array since each FactoredMatrix2D may have * a

* @brief This class represents a factored vector. * * A factored vector is a function over the whole factor space, resulting * from the

* @brief This class implements the Fast Informed Bound algorithm. * * This class is useful in order to obtain a very simple upper bound for

* @brief This class is a generally faster implementation of a Trie. * * This class stores keys in a different way from Trie, which allows it

* @brief This class is used in the Trie in order easily merge id lists. */

* @brief This class implements the GapMin algorithm. * * This method works by repeatedly refining both a lower bound and upper * bound

* @brief This class allows to import generative models from Python. * * This class wraps an instance of a Python class that provides generat