greedy_policy.hpp

Go to the documentation of this file.

#ifndef MLPACK_METHODS_RL_POLICY_GREEDY_POLICY_HPP
#define MLPACK_METHODS_RL_POLICY_GREEDY_POLICY_HPP

#include <mlpack/prereqs.hpp>

namespace mlpack {
namespace rl {

template <typename EnvironmentType>
class GreedyPolicy
{
 public:
  using ActionType = typename EnvironmentType::Action;

  GreedyPolicy(const double initialEpsilon,
               const size_t annealInterval,
               const double minEpsilon) :
      epsilon(initialEpsilon),
      minEpsilon(minEpsilon),
      delta((initialEpsilon - minEpsilon) / annealInterval)
  { /* Nothing to do here. */ }

  ActionType Sample(const arma::colvec& actionValue, bool deterministic = false)
  {
    double exploration = math::Random();

    // Select the action randomly.
    if (!deterministic && exploration < epsilon)
      return static_cast<ActionType>(math::RandInt(ActionType::size));

    // Select the action greedily.
    return static_cast<ActionType>(
        arma::as_scalar(arma::find(actionValue == actionValue.max(), 1)));
  }

  void Anneal()
  {
    epsilon -= delta;
    epsilon = std::max(minEpsilon, epsilon);
  }

  const double& Epsilon() const { return epsilon; }

 private:
  double epsilon;

  double minEpsilon;

  double delta;
};

} // namespace rl
} // namespace mlpack

#endif