Q-learning-in-C/constant.h

18 lines
355 B
C

#define BOT_SYMBOL 1
#define OPPONENT_SYMBOL 2
#define EPSILON 0.9 // Epsilon-greedy
#define LR 0.1 // learning rate
#define LAMBDA 0.9 // discount factor
#define STATE_NUM 19683
#define ACTION_NUM 7
#define EPISODE_NUM 1000000
#define FIRST true
#define ROW_NUM 6
#define COL_NUM 7
#define BIGNUM_LEN 22
#define TABLE_SIZE 1000000000