TopTwoThompsonSamplingPolicy.hpp

Go to the documentation of this file.

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