multi_armed_bandits.h

Go to the documentation of this file.

#ifndef MULTI_ARMED_BANDITS_H
#define MULTI_ARMED_BANDITS_H
 
#include "bitrl/bitrl_types.h"
#include "bitrl/envs/env_base.h"
#include "bitrl/envs/time_step.h"
#include "bitrl/utils/maths/statistics/distributions/bernoulli_dist.h"
 
#include <any>
#include <string>
#include <unordered_map>
#include <vector>
#include <atomic>
 
namespace bitrl
{
namespace envs::bandits
{
 
struct MultiArmedBanditsSpace
{
 
    typedef Null state_space;
 
    typedef Null state_type;
 
    typedef Null action_space;
 
    typedef uint_t action_type;
};
 
class MultiArmedBandits final : public EnvBase<TimeStep<bool>, MultiArmedBanditsSpace>
{
 
public:
    static const std::string name;
 
    typedef EnvBase<TimeStep<bool>, MultiArmedBanditsSpace> base_type;
 
    typedef typename base_type::time_step_type time_step_type;
 
    typedef typename base_type::state_space_type state_space_type;
 
    typedef typename base_type::action_space_type action_space_type;
 
    typedef typename base_type::action_type action_type;
 
    typedef typename base_type::state_type state_type;
 
    MultiArmedBandits();
 
    virtual void
    make(const std::string &version, const std::unordered_map<std::string, std::any> &options,
         const std::unordered_map<std::string, std::any> &reset_options) override final;
 
    virtual void close() override final;
 
    virtual time_step_type reset() override final;
 
    virtual time_step_type step(const action_type &action) override final;
 
    uint_t n_actions() const noexcept { return bandits_.size(); }
 
    real_t success_reward() const noexcept { return success_reward_; }
 
    real_t fail_reward() const noexcept { return fail_reward_; }
 
    static uint_t n_copies() {
        return n_copies_.load();
    }
 
private:
    uint_t seed_;
 
    static std::atomic<uint_t> n_copies_;
 
    real_t success_reward_;
 
    real_t fail_reward_;
 
    std::vector<utils::maths::stats::BernoulliDist> bandits_;
};
 
}
} // namespace bitrl
 
#endif