ThompsonSamplingPolicy.hpp

Go to the documentation of this file.

#ifndef AI_TOOLBOX_BANDIT_THOMPSON_SAMPLING_POLICY_HEADER_FILE
#define AI_TOOLBOX_BANDIT_THOMPSON_SAMPLING_POLICY_HEADER_FILE
 
#include <random>
 
#include <AIToolbox/Bandit/Types.hpp>
#include <AIToolbox/Bandit/Experience.hpp>
#include <AIToolbox/Bandit/Policies/PolicyInterface.hpp>
 
namespace AIToolbox::Bandit {
    class ThompsonSamplingPolicy : public PolicyInterface {
        public:
            ThompsonSamplingPolicy(const Experience & exp);
 
            virtual size_t sampleAction() const override;
 
            virtual double getActionProbability(const size_t & a) const override;
 
            virtual Vector getPolicy() const override;
 
            const Experience & getExperience() const;
 
        private:
            const Experience & exp_;
    };
}
 
#endif