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)
 103 {
 104         if (u->ownermap.playouts < GJ_MINGAMES)
 105                 return false;
 106
 107         struct move_queue mq = { .moves = 0 };
 108         if (!u->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 playouts, %s %s can win with %.2f%% probability",
 170                          n->u.playouts * (u->thread_model == TM_ROOT ? u->threads : 1),
 171                          stone2str(color), coord2sstr(n->coord, b),
 172                          tree_node_get_value(u->t, -1, n->u.value) * 100);
 173                 if (abs(u->t->extra_komi) >= 0.5) {
 174                         sprintf(reply + strlen(reply), ", while self-imposing extra komi %.1f",
 175                                 u->t->extra_komi);
 176                 }
 177                 strcat(reply, ".");
 178                 return reply;
 179         }
 180         return NULL;
 181 }
 182
 183 static void
 184 uct_dead_group_list(struct engine *e, struct board *b, struct move_queue *mq)
 185 {
 186         struct uct *u = e->data;
 187         if (u->pass_all_alive)
 188                 return; // no dead groups
 189
 190         bool mock_state = false;
 191
 192         if (!u->t) {
 193                 /* No state, but we cannot just back out - we might
 194                  * have passed earlier, only assuming some stones are
 195                  * dead, and then re-connected, only to lose counting
 196                  * when all stones are assumed alive. */
 197                 /* Mock up some state and seed the ownermap by few
 198                  * simulations. */
 199                 prepare_move(e, b, S_BLACK); assert(u->t);
 200                 for (int i = 0; i < GJ_MINGAMES; i++)
 201                         uct_playout(u, b, S_BLACK, u->t);
 202                 mock_state = true;
 203         }
 204
 205         dead_group_list(u, b, mq);
 206
 207         if (mock_state) {
 208                 /* Clean up the mock state in case we will receive
 209                  * a genmove; we could get a non-alternating-move
 210                  * error from prepare_move() in that case otherwise. */
 211                 reset_state(u);
 212         }
 213 }
 214
 215 static void
 216 playout_policy_done(struct playout_policy *p)
 217 {
 218         if (p->done) p->done(p);
 219         if (p->data) free(p->data);
 220         free(p);
 221 }
 222
 223 static void
 224 uct_done(struct engine *e)
 225 {
 226         /* This is called on engine reset, especially when clear_board
 227          * is received and new game should begin. */
 228         struct uct *u = e->data;
 229         if (u->t) reset_state(u);
 230         free(u->ownermap.map);
 231
 232         free(u->policy);
 233         free(u->random_policy);
 234         playout_policy_done(u->playout);
 235         uct_prior_done(u->prior);
 236 }
 237
 238
 239 /* Set in main thread in case the playouts should stop. */
 240 volatile sig_atomic_t uct_halt = 0;
 241 /* ID of the running worker thread. */
 242 __thread int thread_id = -1;
 243
 244 static pthread_mutex_t finish_mutex = PTHREAD_MUTEX_INITIALIZER;
 245 static pthread_cond_t finish_cond = PTHREAD_COND_INITIALIZER;
 246 static volatile int finish_thread;
 247 static pthread_mutex_t finish_serializer = PTHREAD_MUTEX_INITIALIZER;
 248
 249 struct spawn_ctx {
 250         int tid;
 251         struct uct *u;
 252         struct board *b;
 253         enum stone color;
 254         struct tree *t;
 255         unsigned long seed;
 256         int games;
 257 };
 258
 259 static void *
 260 spawn_helper(void *ctx_)
 261 {
 262         struct spawn_ctx *ctx = ctx_;
 263         /* Setup */
 264         fast_srandom(ctx->seed);
 265         thread_id = ctx->tid;
 266         /* Run */
 267         ctx->games = uct_playouts(ctx->u, ctx->b, ctx->color, ctx->t, ctx->games);
 268         /* Finish */
 269         pthread_mutex_lock(&finish_serializer);
 270         pthread_mutex_lock(&finish_mutex);
 271         finish_thread = ctx->tid;
 272         pthread_cond_signal(&finish_cond);
 273         pthread_mutex_unlock(&finish_mutex);
 274         return ctx;
 275 }
 276
 277 static int
 278 uct_playouts_parallel(struct uct *u, struct board *b, enum stone color, struct tree *t, int games, bool shared_tree)
 279 {
 280         assert(u->threads > 0);
 281         assert(u->parallel_tree == shared_tree);
 282
 283         int played_games = 0;
 284         pthread_t threads[u->threads];
 285         int joined = 0;
 286
 287         uct_halt = 0;
 288         pthread_mutex_lock(&finish_mutex);
 289         /* Spawn threads... */
 290         for (int ti = 0; ti < u->threads; ti++) {
 291                 struct spawn_ctx *ctx = malloc(sizeof(*ctx));
 292                 ctx->u = u; ctx->b = b; ctx->color = color;
 293                 ctx->t = shared_tree ? t : tree_copy(t);
 294                 ctx->tid = ti; ctx->games = games;
 295                 ctx->seed = fast_random(65536) + ti;
 296                 pthread_create(&threads[ti], NULL, spawn_helper, ctx);
 297                 if (UDEBUGL(2))
 298                         fprintf(stderr, "Spawned thread %d\n", ti);
 299         }
 300
 301         /* ...and collect them back: */
 302         while (joined < u->threads) {
 303                 /* Wait for some thread to finish... */
 304                 pthread_cond_wait(&finish_cond, &finish_mutex);
 305                 /* ...and gather its remnants. */
 306                 struct spawn_ctx *ctx;
 307                 pthread_join(threads[finish_thread], (void **) &ctx);
 308                 played_games += ctx->games;
 309                 joined++;
 310 #ifdef ROOT_PARALLEL
 311                 if (!shared_tree) {
 312                         tree_merge(t, ctx->t);
 313                         tree_done(ctx->t);
 314                 }
 315 #endif
 316                 free(ctx);
 317                 if (UDEBUGL(2))
 318                         fprintf(stderr, "Joined thread %d\n", finish_thread);
 319                 /* Do not get stalled by slow threads. */
 320                 if (joined >= u->threads / 2)
 321                         uct_halt = 1;
 322                 pthread_mutex_unlock(&finish_serializer);
 323         }
 324         pthread_mutex_unlock(&finish_mutex);
 325
 326 #ifdef ROOT_PARALLEL
 327         if (!shared_tree) {
 328                 /* XXX: Should this be done in shared trees as well? */
 329                 tree_normalize(t, u->threads);
 330         }
 331 #endif
 332         return played_games;
 333 }
 334
 335
 336 typedef int (*uct_threaded_playouts)(struct uct *u, struct board *b, enum stone color, struct tree *t, int games);
 337
 338 static int
 339 uct_playouts_none(struct uct *u, struct board *b, enum stone color, struct tree *t, int games)
 340 {
 341         return uct_playouts(u, b, color, t, games);
 342 }
 343
 344 static int
 345 uct_playouts_root(struct uct *u, struct board *b, enum stone color, struct tree *t, int games)
 346 {
 347         return uct_playouts_parallel(u, b, color, t, games, false);
 348 }
 349
 350 static int
 351 uct_playouts_tree(struct uct *u, struct board *b, enum stone color, struct tree *t, int games)
 352 {
 353         return uct_playouts_parallel(u, b, color, t, games, true);
 354 }
 355
 356 static uct_threaded_playouts threaded_playouts[] = {
 357         uct_playouts_none,
 358         uct_playouts_root,
 359         uct_playouts_tree,
 360         uct_playouts_tree,
 361 };
 362
 363
 364 static coord_t *
 365 uct_genmove(struct engine *e, struct board *b, enum stone color)
 366 {
 367         struct uct *u = e->data;
 368
 369         if (b->superko_violation) {
 370                 fprintf(stderr, "!!! WARNING: SUPERKO VIOLATION OCCURED BEFORE THIS MOVE\n");
 371                 fprintf(stderr, "Maybe you play with situational instead of positional superko?\n");
 372                 fprintf(stderr, "I'm going to ignore the violation, but note that I may miss\n");
 373                 fprintf(stderr, "some moves valid under this ruleset because of this.\n");
 374                 b->superko_violation = false;
 375         }
 376
 377         /* Seed the tree. */
 378         prepare_move(e, b, color);
 379         assert(u->t);
 380
 381         /* Run the simulations. */
 382         int games = u->games;
 383         if (u->t->root->children) {
 384                 int delta = u->t->root->u.playouts * 2 / 3;
 385                 if (u->parallel_tree) delta /= u->threads;
 386                 games -= delta;
 387         }
 388         /* else this is highly read-out but dead-end branch of opening book;
 389          * we need to start from scratch; XXX: Maybe actually base the readout
 390          * count based on number of playouts of best node? */
 391         if (games < u->games && UDEBUGL(2))
 392                 fprintf(stderr, "<pre-simulated %d games skipped>\n", u->games - games);
 393
 394         int played_games;
 395         played_games = threaded_playouts[u->thread_model](u, b, color, u->t, games);
 396
 397         if (UDEBUGL(2))
 398                 tree_dump(u->t, u->dumpthres);
 399
 400         /* Choose the best move from the tree. */
 401         struct tree_node *best = u->policy->choose(u->policy, u->t->root, b, color);
 402         if (!best) {
 403                 reset_state(u);
 404                 return coord_copy(pass);
 405         }
 406         if (UDEBUGL(0)) {
 407                 uct_progress_status(u, u->t, color, played_games);
 408         }
 409         if (UDEBUGL(1))
 410                 fprintf(stderr, "*** WINNER is %s (%d,%d) with score %1.4f (%d/%d:%d games)\n",
 411                         coord2sstr(best->coord, b), coord_x(best->coord, b), coord_y(best->coord, b),
 412                         tree_node_get_value(u->t, 1, best->u.value),
 413                         best->u.playouts, u->t->root->u.playouts, played_games);
 414         if (tree_node_get_value(u->t, 1, best->u.value) < u->resign_ratio && !is_pass(best->coord)) {
 415                 reset_state(u);
 416                 return coord_copy(resign);
 417         }
 418
 419         /* If the opponent just passed and we win counting, always
 420          * pass as well. */
 421         if (b->moves > 1 && is_pass(b->last_move.coord)) {
 422                 /* Make sure enough playouts are simulated. */
 423                 while (u->ownermap.playouts < GJ_MINGAMES)
 424                         uct_playout(u, b, color, u->t);
 425                 if (uct_pass_is_safe(u, b, color)) {
 426                         if (UDEBUGL(0))
 427                                 fprintf(stderr, "<Will rather pass, looks safe enough.>\n");
 428                         best->coord = pass;
 429                 }
 430         }
 431
 432         tree_promote_node(u->t, best);
 433         return coord_copy(best->coord);
 434 }
 435
 436
 437 bool
 438 uct_genbook(struct engine *e, struct board *b, enum stone color)
 439 {
 440         struct uct *u = e->data;
 441         if (!u->t) prepare_move(e, b, color);
 442         assert(u->t);
 443
 444         threaded_playouts[u->thread_model](u, b, color, u->t, u->games);
 445
 446         tree_save(u->t, b, u->games / 100);
 447
 448         return true;
 449 }
 450
 451 void
 452 uct_dumpbook(struct engine *e, struct board *b, enum stone color)
 453 {
 454         struct tree *t = tree_init(b, color);
 455         tree_load(t, b);
 456         tree_dump(t, 0);
 457         tree_done(t);
 458 }
 459
 460
 461 struct uct *
 462 uct_state_init(char *arg, struct board *b)
 463 {
 464         struct uct *u = calloc(1, sizeof(struct uct));
 465
 466         u->debug_level = 1;
 467         u->games = MC_GAMES;
 468         u->gamelen = MC_GAMELEN;
 469         u->expand_p = 2;
 470         u->dumpthres = 1000;
 471         u->playout_amaf = true;
 472         u->playout_amaf_nakade = false;
 473         u->amaf_prior = false;
 474
 475         if (board_size(b) - 2 >= 19)
 476                 u->dynkomi = 200;
 477         u->dynkomi_mask = S_BLACK;
 478
 479         u->thread_model = TM_TREEVL;
 480         u->parallel_tree = true;
 481         u->virtual_loss = true;
 482
 483         u->val_scale = 0.02; u->val_points = 20;
 484
 485         if (arg) {
 486                 char *optspec, *next = arg;
 487                 while (*next) {
 488                         optspec = next;
 489                         next += strcspn(next, ",");
 490                         if (*next) { *next++ = 0; } else { *next = 0; }
 491
 492                         char *optname = optspec;
 493                         char *optval = strchr(optspec, '=');
 494                         if (optval) *optval++ = 0;
 495
 496                         if (!strcasecmp(optname, "debug")) {
 497                                 if (optval)
 498                                         u->debug_level = atoi(optval);
 499                                 else
 500                                         u->debug_level++;
 501                         } else if (!strcasecmp(optname, "games") && optval) {
 502                                 u->games = atoi(optval);
 503                         } else if (!strcasecmp(optname, "gamelen") && optval) {
 504                                 u->gamelen = atoi(optval);
 505                         } else if (!strcasecmp(optname, "expand_p") && optval) {
 506                                 u->expand_p = atoi(optval);
 507                         } else if (!strcasecmp(optname, "dumpthres") && optval) {
 508                                 u->dumpthres = atoi(optval);
 509                         } else if (!strcasecmp(optname, "playout_amaf")) {
 510                                 /* Whether to include random playout moves in
 511                                  * AMAF as well. (Otherwise, only tree moves
 512                                  * are included in AMAF. Of course makes sense
 513                                  * only in connection with an AMAF policy.) */
 514                                 /* with-without: 55.5% (+-4.1) */
 515                                 if (optval && *optval == '0')
 516                                         u->playout_amaf = false;
 517                                 else
 518                                         u->playout_amaf = true;
 519                         } else if (!strcasecmp(optname, "playout_amaf_nakade")) {
 520                                 /* Whether to include nakade moves from playouts
 521                                  * in the AMAF statistics; this tends to nullify
 522                                  * the playout_amaf effect by adding too much
 523                                  * noise. */
 524                                 if (optval && *optval == '0')
 525                                         u->playout_amaf_nakade = false;
 526                                 else
 527                                         u->playout_amaf_nakade = true;
 528                         } else if (!strcasecmp(optname, "playout_amaf_cutoff") && optval) {
 529                                 /* Keep only first N% of playout stage AMAF
 530                                  * information. */
 531                                 u->playout_amaf_cutoff = atoi(optval);
 532                         } else if ((!strcasecmp(optname, "policy") || !strcasecmp(optname, "random_policy")) && optval) {
 533                                 char *policyarg = strchr(optval, ':');
 534                                 struct uct_policy **p = !strcasecmp(optname, "policy") ? &u->policy : &u->random_policy;
 535                                 if (policyarg)
 536                                         *policyarg++ = 0;
 537                                 if (!strcasecmp(optval, "ucb1")) {
 538                                         *p = policy_ucb1_init(u, policyarg);
 539                                 } else if (!strcasecmp(optval, "ucb1amaf")) {
 540                                         *p = policy_ucb1amaf_init(u, policyarg);
 541                                 } else {
 542                                         fprintf(stderr, "UCT: Invalid tree policy %s\n", optval);
 543                                         exit(1);
 544                                 }
 545                         } else if (!strcasecmp(optname, "playout") && optval) {
 546                                 char *playoutarg = strchr(optval, ':');
 547                                 if (playoutarg)
 548                                         *playoutarg++ = 0;
 549                                 if (!strcasecmp(optval, "moggy")) {
 550                                         u->playout = playout_moggy_init(playoutarg);
 551                                 } else if (!strcasecmp(optval, "light")) {
 552                                         u->playout = playout_light_init(playoutarg);
 553                                 } else if (!strcasecmp(optval, "elo")) {
 554                                         u->playout = playout_elo_init(playoutarg);
 555                                 } else {
 556                                         fprintf(stderr, "UCT: Invalid playout policy %s\n", optval);
 557                                         exit(1);
 558                                 }
 559                         } else if (!strcasecmp(optname, "prior") && optval) {
 560                                 u->prior = uct_prior_init(optval);
 561                         } else if (!strcasecmp(optname, "amaf_prior") && optval) {
 562                                 u->amaf_prior = atoi(optval);
 563                         } else if (!strcasecmp(optname, "threads") && optval) {
 564                                 u->threads = atoi(optval);
 565                         } else if (!strcasecmp(optname, "thread_model") && optval) {
 566                                 if (!strcasecmp(optval, "none")) {
 567                                         /* Turn off multi-threaded reading. */
 568                                         u->thread_model = TM_NONE;
 569 #ifdef ROOT_PARALLEL
 570                                 } else if (!strcasecmp(optval, "root")) {
 571                                         /* Root parallelization - each thread
 572                                          * does independent search, trees are
 573                                          * merged at the end. */
 574                                         u->thread_model = TM_ROOT;
 575                                         u->parallel_tree = false;
 576                                         u->virtual_loss = false;
 577 #endif
 578                                 } else if (!strcasecmp(optval, "tree")) {
 579                                         /* Tree parallelization - all threads
 580                                          * grind on the same tree. */
 581                                         u->thread_model = TM_TREE;
 582                                         u->parallel_tree = true;
 583                                         u->virtual_loss = false;
 584                                 } else if (!strcasecmp(optval, "treevl")) {
 585                                         /* Tree parallelization, but also
 586                                          * with virtual losses - this discou-
 587                                          * rages most threads choosing the
 588                                          * same tree branches to read. */
 589                                         u->thread_model = TM_TREEVL;
 590                                         u->parallel_tree = true;
 591                                         u->virtual_loss = true;
 592                                 } else {
 593                                         fprintf(stderr, "UCT: Invalid thread model %s\n", optval);
 594                                         exit(1);
 595                                 }
 596                         } else if (!strcasecmp(optname, "force_seed") && optval) {
 597                                 u->force_seed = atoi(optval);
 598                         } else if (!strcasecmp(optname, "no_book")) {
 599                                 u->no_book = true;
 600                         } else if (!strcasecmp(optname, "dynkomi")) {
 601                                 /* Dynamic komi in handicap game; linearly
 602                                  * decreases to basic settings until move
 603                                  * #optval. */
 604                                 u->dynkomi = optval ? atoi(optval) : 150;
 605                         } else if (!strcasecmp(optname, "dynkomi_mask") && optval) {
 606                                 /* Bitmask of colors the player must be
 607                                  * for dynkomi be applied; you may want
 608                                  * to use dynkomi_mask=3 to allow dynkomi
 609                                  * even in games where Pachi is white. */
 610                                 u->dynkomi_mask = atoi(optval);
 611                         } else if (!strcasecmp(optname, "val_scale") && optval) {
 612                                 /* How much of the game result value should be
 613                                  * influenced by win size. Zero means it isn't. */
 614                                 u->val_scale = atof(optval);
 615                         } else if (!strcasecmp(optname, "val_points") && optval) {
 616                                 /* Maximum size of win to be scaled into game
 617                                  * result value. Zero means boardsize^2. */
 618                                 u->val_points = atoi(optval) * 2; // result values are doubled
 619                         } else if (!strcasecmp(optname, "val_extra")) {
 620                                 /* If false, the score coefficient will be simply
 621                                  * added to the value, instead of scaling the result
 622                                  * coefficient because of it. */
 623                                 u->val_extra = !optval || atoi(optval);
 624                         } else if (!strcasecmp(optname, "root_heuristic") && optval) {
 625                                 /* Whether to bias exploration by root node values
 626                                  * (must be supported by the used policy).
 627                                  * 0: Don't.
 628                                  * 1: Do, value = result.
 629                                  * Try to temper the result:
 630                                  * 2: Do, value = 0.5+(result-expected)/2.
 631                                  * 3: Do, value = 0.5+bzz((result-expected)^2). */
 632                                 u->root_heuristic = atoi(optval);
 633                         } else if (!strcasecmp(optname, "pass_all_alive")) {
 634                                 /* Whether to consider all stones alive at the game
 635                                  * end instead of marking dead groupd. */
 636                                 u->pass_all_alive = !optval || atoi(optval);
 637                         } else if (!strcasecmp(optname, "random_policy_chance") && optval) {
 638                                 /* If specified (N), with probability 1/N, random_policy policy
 639                                  * descend is used instead of main policy descend; useful
 640                                  * if specified policy (e.g. UCB1AMAF) can make unduly biased
 641                                  * choices sometimes, you can fall back to e.g.
 642                                  * random_policy=UCB1. */
 643                                 u->random_policy_chance = atoi(optval);
 644                         } else if (!strcasecmp(optname, "banner") && optval) {
 645                                 /* Additional banner string. This must come as the
 646                                  * last engine parameter. */
 647                                 if (*next) *--next = ',';
 648                                 u->banner = strdup(optval);
 649                                 break;
 650                         } else {
 651                                 fprintf(stderr, "uct: Invalid engine argument %s or missing value\n", optname);
 652                                 exit(1);
 653                         }
 654                 }
 655         }
 656
 657         u->resign_ratio = 0.2; /* Resign when most games are lost. */
 658         u->loss_threshold = 0.85; /* Stop reading if after at least 5000 playouts this is best value. */
 659         if (!u->policy)
 660                 u->policy = policy_ucb1amaf_init(u, NULL);
 661         if (!u->threads) {
 662                 u->thread_model = TM_NONE;
 663                 u->parallel_tree = false;
 664         }
 665
 666         if (!!u->random_policy_chance ^ !!u->random_policy) {
 667                 fprintf(stderr, "uct: Only one of random_policy and random_policy_chance is set\n");
 668                 exit(1);
 669         }
 670
 671         if (!u->prior)
 672                 u->prior = uct_prior_init(NULL);
 673
 674         if (!u->playout)
 675                 u->playout = playout_moggy_init(NULL);
 676         u->playout->debug_level = u->debug_level;
 677
 678         u->ownermap.map = malloc(board_size2(b) * sizeof(u->ownermap.map[0]));
 679
 680         /* Some things remain uninitialized for now - the opening book
 681          * is not loaded and the tree not set up. */
 682         /* This will be initialized in setup_state() at the first move
 683          * received/requested. This is because right now we are not aware
 684          * about any komi or handicap setup and such. */
 685
 686         return u;
 687 }
 688
 689 struct engine *
 690 engine_uct_init(char *arg, struct board *b)
 691 {
 692         struct uct *u = uct_state_init(arg, b);
 693         struct engine *e = calloc(1, sizeof(struct engine));
 694         e->name = "UCT Engine";
 695         e->printhook = uct_printhook_ownermap;
 696         e->notify_play = uct_notify_play;
 697         e->chat = uct_chat;
 698         e->genmove = uct_genmove;
 699         e->dead_group_list = uct_dead_group_list;
 700         e->done = uct_done;
 701         e->data = u;
 702
 703         const char banner[] = "I'm playing UCT. When I'm losing, I will resign, "
 704                 "if I think I win, I play until you pass. "
 705                 "Anyone can send me 'winrate' in private chat to get my assessment of the position.";
 706         if (!u->banner) u->banner = "";
 707         e->comment = malloc(sizeof(banner) + strlen(u->banner) + 1);
 708         sprintf(e->comment, "%s %s", banner, u->banner);
 709
 710         return e;
 711 }