LLRPolicy.hpp

Go to the documentation of this file.

#ifndef AI_TOOLBOX_FACTORED_BANDIT_LEARNING_WITH_LINEAR_REWARDS_POLICY_HEADER_FILE
#define AI_TOOLBOX_FACTORED_BANDIT_LEARNING_WITH_LINEAR_REWARDS_POLICY_HEADER_FILE
 
#include <AIToolbox/Factored/Bandit/Types.hpp>
#include <AIToolbox/Factored/Bandit/Experience.hpp>
#include <AIToolbox/Factored/Utils/FilterMap.hpp>
#include <AIToolbox/Factored/Bandit/Policies/PolicyInterface.hpp>
 
namespace AIToolbox::Factored::Bandit {
    class LLRPolicy : public PolicyInterface {
        public:
            LLRPolicy(const Experience & exp);
 
            virtual Action sampleAction() const override;
 
            virtual double getActionProbability(const Action & a) const override;
 
            const Experience & getExperience() const;
 
        private:
            const Experience & exp_;
            unsigned L;
    };
}
 
#endif