Remove backpropagation beta/gamma. (#203)

Author: mooskagh

src/mcts/node.cc

@@ -176,13 +176,9 @@ bool Node::TryStartScoreUpdate() {
 
 void Node::CancelScoreUpdate() { --n_in_flight_; }
 
-void Node::FinalizeScoreUpdate(float v, float gamma, float beta) {
+void Node::FinalizeScoreUpdate(float v) {
   // Recompute Q.
-  if (gamma == 1.0f && beta == 1.0f) {
-    q_ += (v - q_) / (n_ + 1);
-  } else {
-    q_ += (v - q_) / (std::pow(static_cast<float>(n_), gamma) * beta + 1);
-  }
+  q_ += (v - q_) / (n_ + 1);
   // Increment N.
   ++n_;
   // Decrement virtual loss.

src/mcts/node.h

@@ -166,7 +166,7 @@ class Node {
   // * Q (weighted average of all V in a subtree)
   // * N (+=1)
   // * N-in-flight (-=1)
-  void FinalizeScoreUpdate(float v, float gamma, float beta);
+  void FinalizeScoreUpdate(float v);
 
   // Updates max depth, if new depth is larger.
   void UpdateMaxDepth(int depth);
@@ -243,8 +243,12 @@ class EdgeAndNode {
   EdgeAndNode() = default;
   EdgeAndNode(Edge* edge, Node* node) : edge_(edge), node_(node) {}
   explicit operator bool() const { return edge_ != nullptr; }
-  bool operator==(const EdgeAndNode& other) const { return edge_ == other.edge_; }
-  bool operator!=(const EdgeAndNode& other) const { return edge_ != other.edge_; }
+  bool operator==(const EdgeAndNode& other) const {
+    return edge_ == other.edge_;
+  }
+  bool operator!=(const EdgeAndNode& other) const {
+    return edge_ != other.edge_;
+  }
   bool HasNode() const { return node_ != nullptr; }
   Edge* edge() const { return edge_; }
   Node* node() const { return node_; }

src/mcts/search.cc

@@ -47,8 +47,6 @@ const char* Search::kCacheHistoryLengthStr =
 const char* Search::kPolicySoftmaxTempStr = "Policy softmax temperature";
 const char* Search::kAllowedNodeCollisionsStr =
     "Allowed node collisions, per batch";
-const char* Search::kBackPropagateBetaStr = "Backpropagation gamma";
-const char* Search::kBackPropagateGammaStr = "Backpropagation beta";
 
 namespace {
 const int kSmartPruningToleranceNodes = 100;
@@ -79,10 +77,6 @@ void Search::PopulateUciParams(OptionsParser* options) {
                             "policy-softmax-temp") = 1.0f;
   options->Add<IntOption>(kAllowedNodeCollisionsStr, 0, 1024,
                           "allowed-node-collisions") = 0;
-  options->Add<FloatOption>(kBackPropagateBetaStr, 0.0f, 100.0f,
-                            "backpropagate-beta") = 1.0f;
-  options->Add<FloatOption>(kBackPropagateGammaStr, -100.0f, 100.0f,
-                            "backpropagate-gamma") = 1.0f;
 }
 
 Search::Search(const NodeTree& tree, Network* network,
@@ -109,9 +103,7 @@ Search::Search(const NodeTree& tree, Network* network,
       kFpuReduction(options.Get<float>(kFpuReductionStr)),
       kCacheHistoryLength(options.Get<int>(kCacheHistoryLengthStr)),
       kPolicySoftmaxTemp(options.Get<float>(kPolicySoftmaxTempStr)),
-      kAllowedNodeCollisions(options.Get<int>(kAllowedNodeCollisionsStr)),
-      kBackPropagateBeta(options.Get<float>(kBackPropagateBetaStr)),
-      kBackPropagateGamma(options.Get<float>(kBackPropagateGammaStr)) {}
+      kAllowedNodeCollisions(options.Get<int>(kAllowedNodeCollisionsStr)) {}
 
 namespace {
 void ApplyDirichletNoise(Node* node, float eps, double alpha) {
@@ -889,8 +881,7 @@ void SearchWorker::DoBackupUpdate() {
     for (Node* n = node; n != search_->root_node_->GetParent();
          n = n->GetParent()) {
       ++depth;
-      n->FinalizeScoreUpdate(v, search_->kBackPropagateGamma,
-                             search_->kBackPropagateBeta);
+      n->FinalizeScoreUpdate(v);
       // Q will be flipped for opponent.
       v = -v;
 
@@ -903,8 +894,8 @@ void SearchWorker::DoBackupUpdate() {
       // Best move.
       if (n->GetParent() == search_->root_node_ &&
           search_->best_move_edge_.GetN() <= n->GetN()) {
-            search_->best_move_edge_ =
-                search_->GetBestChildNoTemperature(search_->root_node_);
+        search_->best_move_edge_ =
+            search_->GetBestChildNoTemperature(search_->root_node_);
       }
     }
     ++search_->total_playouts_;

src/mcts/search.h

@@ -93,8 +93,6 @@ class Search {
   static const char* kCacheHistoryLengthStr;
   static const char* kPolicySoftmaxTempStr;
   static const char* kAllowedNodeCollisionsStr;
-  static const char* kBackPropagateBetaStr;
-  static const char* kBackPropagateGammaStr;
 
  private:
   // Returns the best move, maybe with temperature (according to the settings).
@@ -167,8 +165,6 @@ class Search {
   const bool kCacheHistoryLength;
   const float kPolicySoftmaxTemp;
   const int kAllowedNodeCollisions;
-  const float kBackPropagateBeta;
-  const float kBackPropagateGamma;
 
   friend class SearchWorker;
 };