From 22027458d13db2494ff76e415a51b100e554dea4 Mon Sep 17 00:00:00 2001
From: Petr Baudis <pasky@ucw.cz>
Date: Sun, 29 Jul 2012 01:04:52 +0200
Subject: [PATCH] Support for value scaling with variable coefficient:
 val_bytemp, val_bytemp_min

---
 uct/internal.h |  2 ++
 uct/uct.c      |  9 +++++++++
 uct/walk.c     | 16 ++++++++++++----
 3 files changed, 23 insertions(+), 4 deletions(-)

diff --git a/uct/internal.h b/uct/internal.h
index ae25701..62bb934 100644
--- a/uct/internal.h
+++ b/uct/internal.h
@@ -80,6 +80,8 @@ struct uct {
 	int val_points;
 	bool val_extra;
 	bool val_byavg;
+	bool val_bytemp;
+	floating_t val_bytemp_min;
 
 	int random_policy_chance;
 	bool local_tree;
diff --git a/uct/uct.c b/uct/uct.c
index 4aac18c..d1ebe25 100644
--- a/uct/uct.c
+++ b/uct/uct.c
@@ -911,6 +911,15 @@ uct_state_init(char *arg, struct board *b)
 				 * be relative to average score in the current
 				 * search episode inst. of jigo. */
 				u->val_byavg = !optval || atoi(optval);
+			} else if (!strcasecmp(optname, "val_bytemp")) {
+				/* If true, the value scaling coefficient
+				 * is different based on value extremity
+				 * (dist. from 0.5), linear between
+				 * val_bytemp_min, val_scale. */
+				u->val_bytemp = !optval || atoi(optval);
+			} else if (!strcasecmp(optname, "val_bytemp_min") && optval) {
+				/* Minimum val_scale in case of val_bytemp. */
+				u->val_bytemp_min = atof(optval);
 
 			/** Local trees */
 			/* (Purely experimental. Does not work - yet!) */
diff --git a/uct/walk.c b/uct/walk.c
index 278bc25..2ad2c63 100644
--- a/uct/walk.c
+++ b/uct/walk.c
@@ -268,7 +268,7 @@ uct_leaf_node(struct uct *u, struct board *b, enum stone player_color,
 }
 
 static floating_t
-scale_value(struct uct *u, struct board *b, int result)
+scale_value(struct uct *u, struct board *b, enum stone node_color, struct tree_node *significant[2], int result)
 {
 	floating_t rval = result > 0 ? 1.0 : result < 0 ? 0.0 : 0.5;
 	if (u->val_scale && result != 0) {
@@ -278,6 +278,14 @@ scale_value(struct uct *u, struct board *b, int result)
 			result -= u->t->avg_score.value * 2;
 		}
 
+		double scale = u->val_scale;
+		if (u->val_bytemp) {
+			/* xvalue is 0 at 0.5, 1 at 0 or 1 */
+			/* No correction for parity necessary. */
+			double xvalue = significant[node_color] ? fabs(significant[node_color]->u.value - 0.5) * 2 : 0;
+			scale = u->val_bytemp_min + (u->val_scale - u->val_bytemp_min) * xvalue;
+		}
+
 		int vp = u->val_points;
 		if (!vp) {
 			vp = board_size(b) - 1; vp *= vp; vp *= 2;
@@ -287,9 +295,9 @@ scale_value(struct uct *u, struct board *b, int result)
 		sval = sval > 1 ? 1 : sval;
 		if (result < 0) sval = 1 - sval;
 		if (u->val_extra)
-			rval += u->val_scale * sval;
+			rval += scale * sval;
 		else
-			rval = (1 - u->val_scale) * rval + u->val_scale * sval;
+			rval = (1 - scale) * rval + scale * sval;
 		// fprintf(stderr, "score %d => sval %f, rval %f\n", result, sval, rval);
 	}
 	return rval;
@@ -557,7 +565,7 @@ uct_playout(struct uct *u, struct board *b, enum stone player_color, struct tree
 	/* Record the result. */
 
 	assert(n == t->root || n->parent);
-	floating_t rval = scale_value(u, b, result);
+	floating_t rval = scale_value(u, b, node_color, significant, result);
 	u->policy->update(u->policy, t, n, node_color, player_color, &amaf, &b2, rval);
 
 	stats_add_result(&t->avg_score, result / 2, 1);
-- 
2.11.4.GIT