uct/uct.c

   1 #include <assert.h>
   2 #include <pthread.h>
   3 #include <signal.h>
   4 #include <stdio.h>
   5 #include <stdlib.h>
   6 #include <string.h>
   7
   8 #define DEBUG
   9
  10 #include "debug.h"
  11 #include "board.h"
  12 #include "gtp.h"
  13 #include "move.h"
  14 #include "mq.h"
  15 #include "playout.h"
  16 #include "playout/elo.h"
  17 #include "playout/moggy.h"
  18 #include "playout/light.h"
  19 #include "random.h"
  20 #include "tactics.h"
  21 #include "uct/internal.h"
  22 #include "uct/prior.h"
  23 #include "uct/tree.h"
  24 #include "uct/uct.h"
  25 #include "uct/walk.h"
  26
  27 struct uct_policy *policy_ucb1_init(struct uct *u, char *arg);
  28 struct uct_policy *policy_ucb1amaf_init(struct uct *u, char *arg);
  29
  30
  31 #define MC_GAMES        80000
  32 #define MC_GAMELEN      MAX_GAMELEN
  33
  34 /* How big proportion of ownermap counts must be of one color to consider
  35  * the point sure. */
  36 #define GJ_THRES        0.8
  37 /* How many games to consider at minimum before judging groups. */
  38 #define GJ_MINGAMES     500
  39
  40
  41 static void
  42 setup_state(struct uct *u, struct board *b, enum stone color)
  43 {
  44         u->t = tree_init(b, color);
  45         if (u->force_seed)
  46                 fast_srandom(u->force_seed);
  47         if (UDEBUGL(0))
  48                 fprintf(stderr, "Fresh board with random seed %lu\n", fast_getseed());
  49         //board_print(b, stderr);
  50         if (!u->no_book && b->moves == 0) {
  51                 assert(color == S_BLACK);
  52                 tree_load(u->t, b);
  53         }
  54 }
  55
  56 static void
  57 reset_state(struct uct *u)
  58 {
  59         assert(u->t);
  60         tree_done(u->t); u->t = NULL;
  61 }
  62
  63 static void
  64 prepare_move(struct engine *e, struct board *b, enum stone color)
  65 {
  66         struct uct *u = e->data;
  67
  68         if (u->t) {
  69                 /* Verify that we have sane state. */
  70                 assert(b->es == u);
  71                 assert(u->t && b->moves);
  72                 if (color != stone_other(u->t->root_color)) {
  73                         fprintf(stderr, "Fatal: Non-alternating play detected %d %d\n",
  74                                 color, u->t->root_color);
  75                         exit(1);
  76                 }
  77
  78         } else {
  79                 /* We need fresh state. */
  80                 b->es = u;
  81                 setup_state(u, b, color);
  82         }
  83
  84         if (u->dynkomi && u->dynkomi > b->moves && (color & u->dynkomi_mask))
  85                 u->t->extra_komi = uct_get_extra_komi(u, b);
  86
  87         u->ownermap.playouts = 0;
  88         memset(u->ownermap.map, 0, board_size2(b) * sizeof(u->ownermap.map[0]));
  89 }
  90
  91 static void
  92 dead_group_list(struct uct *u, struct board *b, struct move_queue *mq)
  93 {
  94         struct group_judgement gj;
  95         gj.thres = GJ_THRES;
  96         gj.gs = alloca(board_size2(b) * sizeof(gj.gs[0]));
  97         board_ownermap_judge_group(b, &u->ownermap, &gj);
  98         groups_of_status(b, &gj, GS_DEAD, mq);
  99 }
 100
 101 bool
 102 uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive)
 103 {
 104         if (u->ownermap.playouts < GJ_MINGAMES)
 105                 return false;
 106
 107         struct move_queue mq = { .moves = 0 };
 108         if (!pass_all_alive)
 109                 dead_group_list(u, b, &mq);
 110         return pass_is_safe(b, color, &mq);
 111 }
 112
 113
 114 static void
 115 uct_printhook_ownermap(struct board *board, coord_t c, FILE *f)
 116 {
 117         struct uct *u = board->es;
 118         assert(u);
 119         const char chr[] = ":XO,"; // dame, black, white, unclear
 120         const char chm[] = ":xo,";
 121         char ch = chr[board_ownermap_judge_point(&u->ownermap, c, GJ_THRES)];
 122         if (ch == ',') { // less precise estimate then?
 123                 ch = chm[board_ownermap_judge_point(&u->ownermap, c, 0.67)];
 124         }
 125         fprintf(f, "%c ", ch);
 126 }
 127
 128 static char *
 129 uct_notify_play(struct engine *e, struct board *b, struct move *m)
 130 {
 131         struct uct *u = e->data;
 132         if (!u->t) {
 133                 /* No state, create one - this is probably game beginning
 134                  * and we need to load the opening book right now. */
 135                 prepare_move(e, b, m->color);
 136                 assert(u->t);
 137         }
 138
 139         if (is_resign(m->coord)) {
 140                 /* Reset state. */
 141                 reset_state(u);
 142                 return NULL;
 143         }
 144
 145         /* Promote node of the appropriate move to the tree root. */
 146         assert(u->t->root);
 147         if (!tree_promote_at(u->t, b, m->coord)) {
 148                 if (UDEBUGL(0))
 149                         fprintf(stderr, "Warning: Cannot promote move node! Several play commands in row?\n");
 150                 reset_state(u);
 151                 return NULL;
 152         }
 153
 154         return NULL;
 155 }
 156
 157 static char *
 158 uct_chat(struct engine *e, struct board *b, char *cmd)
 159 {
 160         struct uct *u = e->data;
 161         static char reply[1024];
 162
 163         cmd += strspn(cmd, " \n\t");
 164         if (!strncasecmp(cmd, "winrate", 7)) {
 165                 if (!u->t)
 166                         return "no game context (yet?)";
 167                 enum stone color = u->t->root_color;
 168                 struct tree_node *n = u->t->root;
 169                 snprintf(reply, 1024, "In %d*%d playouts, %s %s can win with %.2f%% probability",
 170                          n->u.playouts, u->threads, stone2str(color), coord2sstr(n->coord, b),
 171                          tree_node_get_value(u->t, -1, n->u.value) * 100);
 172                 if (abs(u->t->extra_komi) >= 0.5) {
 173                         sprintf(reply + strlen(reply), ", while self-imposing extra komi %.1f",
 174                                 u->t->extra_komi);
 175                 }
 176                 strcat(reply, ".");
 177                 return reply;
 178         }
 179         return NULL;
 180 }
 181
 182 static void
 183 uct_dead_group_list(struct engine *e, struct board *b, struct move_queue *mq)
 184 {
 185         struct uct *u = e->data;
 186         if (u->pass_all_alive)
 187                 return; // no dead groups
 188
 189         bool mock_state = false;
 190
 191         if (!u->t) {
 192                 /* No state, but we cannot just back out - we might
 193                  * have passed earlier, only assuming some stones are
 194                  * dead, and then re-connected, only to lose counting
 195                  * when all stones are assumed alive. */
 196                 /* Mock up some state and seed the ownermap by few
 197                  * simulations. */
 198                 prepare_move(e, b, S_BLACK); assert(u->t);
 199                 for (int i = 0; i < GJ_MINGAMES; i++)
 200                         uct_playout(u, b, S_BLACK, u->t);
 201                 mock_state = true;
 202         }
 203
 204         dead_group_list(u, b, mq);
 205
 206         if (mock_state) {
 207                 /* Clean up the mock state in case we will receive
 208                  * a genmove; we could get a non-alternating-move
 209                  * error from prepare_move() in that case otherwise. */
 210                 reset_state(u);
 211         }
 212 }
 213
 214 static void
 215 playout_policy_done(struct playout_policy *p)
 216 {
 217         if (p->done) p->done(p);
 218         if (p->data) free(p->data);
 219         free(p);
 220 }
 221
 222 static void
 223 uct_done(struct engine *e)
 224 {
 225         /* This is called on engine reset, especially when clear_board
 226          * is received and new game should begin. */
 227         struct uct *u = e->data;
 228         if (u->t) reset_state(u);
 229         free(u->ownermap.map);
 230
 231         free(u->policy);
 232         free(u->random_policy);
 233         playout_policy_done(u->playout);
 234         uct_prior_done(u->prior);
 235 }
 236
 237
 238 /* Set in main thread in case the playouts should stop. */
 239 volatile sig_atomic_t uct_halt = 0;
 240 /* ID of the running worker thread. */
 241 __thread int thread_id = -1;
 242
 243 static pthread_mutex_t finish_mutex = PTHREAD_MUTEX_INITIALIZER;
 244 static pthread_cond_t finish_cond = PTHREAD_COND_INITIALIZER;
 245 static volatile int finish_thread;
 246 static pthread_mutex_t finish_serializer = PTHREAD_MUTEX_INITIALIZER;
 247
 248 struct spawn_ctx {
 249         int tid;
 250         struct uct *u;
 251         struct board *b;
 252         enum stone color;
 253         struct tree *t;
 254         unsigned long seed;
 255         int games;
 256 };
 257
 258 static void *
 259 spawn_helper(void *ctx_)
 260 {
 261         struct spawn_ctx *ctx = ctx_;
 262         /* Setup */
 263         fast_srandom(ctx->seed);
 264         thread_id = ctx->tid;
 265         /* Run */
 266         ctx->games = uct_playouts(ctx->u, ctx->b, ctx->color, ctx->t, ctx->games);
 267         /* Finish */
 268         pthread_mutex_lock(&finish_serializer);
 269         pthread_mutex_lock(&finish_mutex);
 270         finish_thread = ctx->tid;
 271         pthread_cond_signal(&finish_cond);
 272         pthread_mutex_unlock(&finish_mutex);
 273         return ctx;
 274 }
 275
 276 static int
 277 uct_playouts_parallel(struct uct *u, struct board *b, enum stone color, struct tree *t, int games, bool shared_tree)
 278 {
 279         assert(u->threads > 0);
 280         assert(u->parallel_tree == shared_tree);
 281
 282         int played_games = 0;
 283         pthread_t threads[u->threads];
 284         int joined = 0;
 285
 286         uct_halt = 0;
 287         pthread_mutex_lock(&finish_mutex);
 288         /* Spawn threads... */
 289         for (int ti = 0; ti < u->threads; ti++) {
 290                 struct spawn_ctx *ctx = malloc(sizeof(*ctx));
 291                 ctx->u = u; ctx->b = b; ctx->color = color;
 292                 ctx->t = shared_tree ? t : tree_copy(t);
 293                 ctx->tid = ti; ctx->games = games;
 294                 ctx->seed = fast_random(65536) + ti;
 295                 pthread_create(&threads[ti], NULL, spawn_helper, ctx);
 296                 if (UDEBUGL(2))
 297                         fprintf(stderr, "Spawned thread %d\n", ti);
 298         }
 299
 300         /* ...and collect them back: */
 301         while (joined < u->threads) {
 302                 /* Wait for some thread to finish... */
 303                 pthread_cond_wait(&finish_cond, &finish_mutex);
 304                 /* ...and gather its remnants. */
 305                 struct spawn_ctx *ctx;
 306                 pthread_join(threads[finish_thread], (void **) &ctx);
 307                 played_games += ctx->games;
 308                 joined++;
 309                 if (!shared_tree) {
 310                         tree_merge(t, ctx->t);
 311                         tree_done(ctx->t);
 312                 }
 313                 free(ctx);
 314                 if (UDEBUGL(2))
 315                         fprintf(stderr, "Joined thread %d\n", finish_thread);
 316                 /* Do not get stalled by slow threads. */
 317                 if (joined >= u->threads / 2)
 318                         uct_halt = 1;
 319                 pthread_mutex_unlock(&finish_serializer);
 320         }
 321         pthread_mutex_unlock(&finish_mutex);
 322
 323         if (!shared_tree) {
 324                 /* XXX: Should this be done in shared trees as well? */
 325                 tree_normalize(t, u->threads);
 326         }
 327         return played_games;
 328 }
 329
 330
 331 typedef int (*uct_threaded_playouts)(struct uct *u, struct board *b, enum stone color, struct tree *t, int games);
 332
 333 static int
 334 uct_playouts_none(struct uct *u, struct board *b, enum stone color, struct tree *t, int games)
 335 {
 336         return uct_playouts(u, b, color, t, games);
 337 }
 338
 339 static int
 340 uct_playouts_root(struct uct *u, struct board *b, enum stone color, struct tree *t, int games)
 341 {
 342         return uct_playouts_parallel(u, b, color, t, games, false);
 343 }
 344
 345 static int
 346 uct_playouts_tree(struct uct *u, struct board *b, enum stone color, struct tree *t, int games)
 347 {
 348         return uct_playouts_parallel(u, b, color, t, games, true);
 349 }
 350
 351 static uct_threaded_playouts threaded_playouts[] = {
 352         uct_playouts_none,
 353         uct_playouts_root,
 354         uct_playouts_tree,
 355         uct_playouts_tree,
 356 };
 357
 358
 359 static coord_t *
 360 uct_genmove(struct engine *e, struct board *b, enum stone color, bool pass_all_alive)
 361 {
 362         struct uct *u = e->data;
 363
 364         if (b->superko_violation) {
 365                 fprintf(stderr, "!!! WARNING: SUPERKO VIOLATION OCCURED BEFORE THIS MOVE\n");
 366                 fprintf(stderr, "Maybe you play with situational instead of positional superko?\n");
 367                 fprintf(stderr, "I'm going to ignore the violation, but note that I may miss\n");
 368                 fprintf(stderr, "some moves valid under this ruleset because of this.\n");
 369                 b->superko_violation = false;
 370         }
 371
 372         /* Seed the tree. */
 373         prepare_move(e, b, color);
 374         assert(u->t);
 375
 376         /* Run the simulations. */
 377         int games = u->games;
 378         if (u->t->root->children) {
 379                 int delta = u->t->root->u.playouts * 2 / 3;
 380                 if (u->parallel_tree) delta /= u->threads;
 381                 games -= delta;
 382         }
 383         /* else this is highly read-out but dead-end branch of opening book;
 384          * we need to start from scratch; XXX: Maybe actually base the readout
 385          * count based on number of playouts of best node? */
 386         if (games < u->games && UDEBUGL(2))
 387                 fprintf(stderr, "<pre-simulated %d games skipped>\n", u->games - games);
 388
 389         int played_games;
 390         played_games = threaded_playouts[u->thread_model](u, b, color, u->t, games);
 391
 392         if (UDEBUGL(2))
 393                 tree_dump(u->t, u->dumpthres);
 394
 395         /* Choose the best move from the tree. */
 396         struct tree_node *best = u->policy->choose(u->policy, u->t->root, b, color);
 397         if (!best) {
 398                 reset_state(u);
 399                 return coord_copy(pass);
 400         }
 401         if (UDEBUGL(0)) {
 402                 uct_progress_status(u, u->t, color, played_games);
 403         }
 404         if (UDEBUGL(1))
 405                 fprintf(stderr, "*** WINNER is %s (%d,%d) with score %1.4f (%d/%d:%d games)\n",
 406                         coord2sstr(best->coord, b), coord_x(best->coord, b), coord_y(best->coord, b),
 407                         tree_node_get_value(u->t, 1, best->u.value),
 408                         best->u.playouts, u->t->root->u.playouts, played_games);
 409         if (tree_node_get_value(u->t, 1, best->u.value) < u->resign_ratio && !is_pass(best->coord)) {
 410                 reset_state(u);
 411                 return coord_copy(resign);
 412         }
 413
 414         /* If the opponent just passed and we win counting, always
 415          * pass as well. */
 416         if (b->moves > 1 && is_pass(b->last_move.coord)) {
 417                 /* Make sure enough playouts are simulated. */
 418                 while (u->ownermap.playouts < GJ_MINGAMES)
 419                         uct_playout(u, b, color, u->t);
 420                 if (uct_pass_is_safe(u, b, color, u->pass_all_alive || pass_all_alive)) {
 421                         if (UDEBUGL(0))
 422                                 fprintf(stderr, "<Will rather pass, looks safe enough.>\n");
 423                         best->coord = pass;
 424                 }
 425         }
 426
 427         tree_promote_node(u->t, best);
 428         return coord_copy(best->coord);
 429 }
 430
 431
 432 bool
 433 uct_genbook(struct engine *e, struct board *b, enum stone color)
 434 {
 435         struct uct *u = e->data;
 436         if (!u->t) prepare_move(e, b, color);
 437         assert(u->t);
 438
 439         threaded_playouts[u->thread_model](u, b, color, u->t, u->games);
 440
 441         tree_save(u->t, b, u->games / 100);
 442
 443         return true;
 444 }
 445
 446 void
 447 uct_dumpbook(struct engine *e, struct board *b, enum stone color)
 448 {
 449         struct tree *t = tree_init(b, color);
 450         tree_load(t, b);
 451         tree_dump(t, 0);
 452         tree_done(t);
 453 }
 454
 455
 456 struct uct *
 457 uct_state_init(char *arg, struct board *b)
 458 {
 459         struct uct *u = calloc(1, sizeof(struct uct));
 460
 461         u->debug_level = 1;
 462         u->games = MC_GAMES;
 463         u->gamelen = MC_GAMELEN;
 464         u->expand_p = 2;
 465         u->dumpthres = 1000;
 466         u->playout_amaf = true;
 467         u->playout_amaf_nakade = false;
 468         u->amaf_prior = false;
 469
 470         if (board_size(b) - 2 >= 19)
 471                 u->dynkomi = 200;
 472         u->dynkomi_mask = S_BLACK;
 473
 474         u->thread_model = TM_TREEVL;
 475         u->parallel_tree = true;
 476         u->virtual_loss = true;
 477
 478         u->val_scale = 0.02; u->val_points = 20;
 479
 480         if (arg) {
 481                 char *optspec, *next = arg;
 482                 while (*next) {
 483                         optspec = next;
 484                         next += strcspn(next, ",");
 485                         if (*next) { *next++ = 0; } else { *next = 0; }
 486
 487                         char *optname = optspec;
 488                         char *optval = strchr(optspec, '=');
 489                         if (optval) *optval++ = 0;
 490
 491                         if (!strcasecmp(optname, "debug")) {
 492                                 if (optval)
 493                                         u->debug_level = atoi(optval);
 494                                 else
 495                                         u->debug_level++;
 496                         } else if (!strcasecmp(optname, "games") && optval) {
 497                                 u->games = atoi(optval);
 498                         } else if (!strcasecmp(optname, "gamelen") && optval) {
 499                                 u->gamelen = atoi(optval);
 500                         } else if (!strcasecmp(optname, "expand_p") && optval) {
 501                                 u->expand_p = atoi(optval);
 502                         } else if (!strcasecmp(optname, "dumpthres") && optval) {
 503                                 u->dumpthres = atoi(optval);
 504                         } else if (!strcasecmp(optname, "playout_amaf")) {
 505                                 /* Whether to include random playout moves in
 506                                  * AMAF as well. (Otherwise, only tree moves
 507                                  * are included in AMAF. Of course makes sense
 508                                  * only in connection with an AMAF policy.) */
 509                                 /* with-without: 55.5% (+-4.1) */
 510                                 if (optval && *optval == '0')
 511                                         u->playout_amaf = false;
 512                                 else
 513                                         u->playout_amaf = true;
 514                         } else if (!strcasecmp(optname, "playout_amaf_nakade")) {
 515                                 /* Whether to include nakade moves from playouts
 516                                  * in the AMAF statistics; this tends to nullify
 517                                  * the playout_amaf effect by adding too much
 518                                  * noise. */
 519                                 if (optval && *optval == '0')
 520                                         u->playout_amaf_nakade = false;
 521                                 else
 522                                         u->playout_amaf_nakade = true;
 523                         } else if (!strcasecmp(optname, "playout_amaf_cutoff") && optval) {
 524                                 /* Keep only first N% of playout stage AMAF
 525                                  * information. */
 526                                 u->playout_amaf_cutoff = atoi(optval);
 527                         } else if ((!strcasecmp(optname, "policy") || !strcasecmp(optname, "random_policy")) && optval) {
 528                                 char *policyarg = strchr(optval, ':');
 529                                 struct uct_policy **p = !strcasecmp(optname, "policy") ? &u->policy : &u->random_policy;
 530                                 if (policyarg)
 531                                         *policyarg++ = 0;
 532                                 if (!strcasecmp(optval, "ucb1")) {
 533                                         *p = policy_ucb1_init(u, policyarg);
 534                                 } else if (!strcasecmp(optval, "ucb1amaf")) {
 535                                         *p = policy_ucb1amaf_init(u, policyarg);
 536                                 } else {
 537                                         fprintf(stderr, "UCT: Invalid tree policy %s\n", optval);
 538                                         exit(1);
 539                                 }
 540                         } else if (!strcasecmp(optname, "playout") && optval) {
 541                                 char *playoutarg = strchr(optval, ':');
 542                                 if (playoutarg)
 543                                         *playoutarg++ = 0;
 544                                 if (!strcasecmp(optval, "moggy")) {
 545                                         u->playout = playout_moggy_init(playoutarg);
 546                                 } else if (!strcasecmp(optval, "light")) {
 547                                         u->playout = playout_light_init(playoutarg);
 548                                 } else if (!strcasecmp(optval, "elo")) {
 549                                         u->playout = playout_elo_init(playoutarg);
 550                                 } else {
 551                                         fprintf(stderr, "UCT: Invalid playout policy %s\n", optval);
 552                                         exit(1);
 553                                 }
 554                         } else if (!strcasecmp(optname, "prior") && optval) {
 555                                 u->prior = uct_prior_init(optval);
 556                         } else if (!strcasecmp(optname, "amaf_prior") && optval) {
 557                                 u->amaf_prior = atoi(optval);
 558                         } else if (!strcasecmp(optname, "threads") && optval) {
 559                                 u->threads = atoi(optval);
 560                         } else if (!strcasecmp(optname, "thread_model") && optval) {
 561                                 if (!strcasecmp(optval, "none")) {
 562                                         /* Turn off multi-threaded reading. */
 563                                         u->thread_model = TM_NONE;
 564                                 } else if (!strcasecmp(optval, "root")) {
 565                                         /* Root parallelization - each thread
 566                                          * does independent search, trees are
 567                                          * merged at the end. */
 568                                         u->thread_model = TM_ROOT;
 569                                         u->parallel_tree = false;
 570                                         u->virtual_loss = false;
 571                                 } else if (!strcasecmp(optval, "tree")) {
 572                                         /* Tree parallelization - all threads
 573                                          * grind on the same tree. */
 574                                         u->thread_model = TM_TREE;
 575                                         u->parallel_tree = true;
 576                                         u->virtual_loss = false;
 577                                 } else if (!strcasecmp(optval, "treevl")) {
 578                                         /* Tree parallelization, but also
 579                                          * with virtual losses - this discou-
 580                                          * rages most threads choosing the
 581                                          * same tree branches to read. */
 582                                         u->thread_model = TM_TREEVL;
 583                                         u->parallel_tree = true;
 584                                         u->virtual_loss = true;
 585                                 } else {
 586                                         fprintf(stderr, "UCT: Invalid thread model %s\n", optval);
 587                                         exit(1);
 588                                 }
 589                         } else if (!strcasecmp(optname, "force_seed") && optval) {
 590                                 u->force_seed = atoi(optval);
 591                         } else if (!strcasecmp(optname, "no_book")) {
 592                                 u->no_book = true;
 593                         } else if (!strcasecmp(optname, "dynkomi")) {
 594                                 /* Dynamic komi in handicap game; linearly
 595                                  * decreases to basic settings until move
 596                                  * #optval. */
 597                                 u->dynkomi = optval ? atoi(optval) : 150;
 598                         } else if (!strcasecmp(optname, "dynkomi_mask") && optval) {
 599                                 /* Bitmask of colors the player must be
 600                                  * for dynkomi be applied; you may want
 601                                  * to use dynkomi_mask=3 to allow dynkomi
 602                                  * even in games where Pachi is white. */
 603                                 u->dynkomi_mask = atoi(optval);
 604                         } else if (!strcasecmp(optname, "val_scale") && optval) {
 605                                 /* How much of the game result value should be
 606                                  * influenced by win size. Zero means it isn't. */
 607                                 u->val_scale = atof(optval);
 608                         } else if (!strcasecmp(optname, "val_points") && optval) {
 609                                 /* Maximum size of win to be scaled into game
 610                                  * result value. Zero means boardsize^2. */
 611                                 u->val_points = atoi(optval) * 2; // result values are doubled
 612                         } else if (!strcasecmp(optname, "val_extra")) {
 613                                 /* If false, the score coefficient will be simply
 614                                  * added to the value, instead of scaling the result
 615                                  * coefficient because of it. */
 616                                 u->val_extra = !optval || atoi(optval);
 617                         } else if (!strcasecmp(optname, "root_heuristic") && optval) {
 618                                 /* Whether to bias exploration by root node values
 619                                  * (must be supported by the used policy).
 620                                  * 0: Don't.
 621                                  * 1: Do, value = result.
 622                                  * Try to temper the result:
 623                                  * 2: Do, value = 0.5+(result-expected)/2.
 624                                  * 3: Do, value = 0.5+bzz((result-expected)^2). */
 625                                 u->root_heuristic = atoi(optval);
 626                         } else if (!strcasecmp(optname, "pass_all_alive")) {
 627                                 /* Whether to consider all stones alive at the game
 628                                  * end instead of marking dead groupd. */
 629                                 u->pass_all_alive = !optval || atoi(optval);
 630                         } else if (!strcasecmp(optname, "random_policy_chance") && optval) {
 631                                 /* If specified (N), with probability 1/N, random_policy policy
 632                                  * descend is used instead of main policy descend; useful
 633                                  * if specified policy (e.g. UCB1AMAF) can make unduly biased
 634                                  * choices sometimes, you can fall back to e.g.
 635                                  * random_policy=UCB1. */
 636                                 u->random_policy_chance = atoi(optval);
 637                         } else if (!strcasecmp(optname, "banner") && optval) {
 638                                 /* Additional banner string. This must come as the
 639                                  * last engine parameter. */
 640                                 if (*next) *--next = ',';
 641                                 u->banner = strdup(optval);
 642                                 break;
 643                         } else {
 644                                 fprintf(stderr, "uct: Invalid engine argument %s or missing value\n", optname);
 645                                 exit(1);
 646                         }
 647                 }
 648         }
 649
 650         u->resign_ratio = 0.2; /* Resign when most games are lost. */
 651         u->loss_threshold = 0.85; /* Stop reading if after at least 5000 playouts this is best value. */
 652         if (!u->policy)
 653                 u->policy = policy_ucb1amaf_init(u, NULL);
 654         if (!u->threads) {
 655                 u->thread_model = TM_NONE;
 656                 u->parallel_tree = false;
 657         }
 658
 659         if (!!u->random_policy_chance ^ !!u->random_policy) {
 660                 fprintf(stderr, "uct: Only one of random_policy and random_policy_chance is set\n");
 661                 exit(1);
 662         }
 663
 664         if (!u->prior)
 665                 u->prior = uct_prior_init(NULL);
 666
 667         if (!u->playout)
 668                 u->playout = playout_moggy_init(NULL);
 669         u->playout->debug_level = u->debug_level;
 670
 671         u->ownermap.map = malloc(board_size2(b) * sizeof(u->ownermap.map[0]));
 672
 673         /* Some things remain uninitialized for now - the opening book
 674          * is not loaded and the tree not set up. */
 675         /* This will be initialized in setup_state() at the first move
 676          * received/requested. This is because right now we are not aware
 677          * about any komi or handicap setup and such. */
 678
 679         return u;
 680 }
 681
 682 struct engine *
 683 engine_uct_init(char *arg, struct board *b)
 684 {
 685         struct uct *u = uct_state_init(arg, b);
 686         struct engine *e = calloc(1, sizeof(struct engine));
 687         e->name = "UCT Engine";
 688         e->printhook = uct_printhook_ownermap;
 689         e->notify_play = uct_notify_play;
 690         e->chat = uct_chat;
 691         e->genmove = uct_genmove;
 692         e->dead_group_list = uct_dead_group_list;
 693         e->done = uct_done;
 694         e->data = u;
 695
 696         const char banner[] = "I'm playing UCT. When I'm losing, I will resign, "
 697                 "if I think I win, I play until you pass. "
 698                 "Anyone can send me 'winrate' in private chat to get my assessment of the position.";
 699         if (!u->banner) u->banner = "";
 700         e->comment = malloc(sizeof(banner) + strlen(u->banner) + 1);
 701         sprintf(e->comment, "%s %s", banner, u->banner);
 702
 703         return e;
 704 }