20 void update(
string arm,
float reward);
BanditOperator(vector< string > arms)
Constructs a BanditOperator object with a vector of arms.
string choose()
Chooses an arm based on the given tree and fitness. Should call sample_probs internally.
std::map< string, float > sample_probs(bool update)
Samples the probabilities of the arms.
~ThompsonSamplingBandit()
void update(string arm, float reward)
Updates the reward for a specific arm.
std::map< string, float > alphas
ThompsonSamplingBandit(vector< string > arms, bool dynamic=false)
std::map< string, float > betas
< nsga2 selection operator for getting the front