nonuct_cnode_patch.diff

diff --git a/core/mcts/ctree/ctree_sampled/lib/cnode.cpp b/core/mcts/ctree/ctree_sampled/lib/cnode.cpp
index 16e8a14..a322972 100644
--- a/core/mcts/ctree/ctree_sampled/lib/cnode.cpp
+++ b/core/mcts/ctree/ctree_sampled/lib/cnode.cpp
@@ -330,7 +330,16 @@ namespace tree
         if (value_score > 1)
             value_score = 1;
 
-        float ucb_value = prior_score + value_score;
+        // NonUCT: additive surrogate term. this->nonuct_weight defaults to
+        // 0.0 (set via CTree constructor / a new setter), so with no
+        // wiring from Python this reduces exactly to the original MAZero
+        // pUCT score -- existing configs/results are unaffected unless the
+        // weight is explicitly turned on. child->nonuct_score is raw
+        // eta(theta,a) (unbounded), so it is squashed to comparable scale
+        // via the same asinh-family shape rather than mixed in raw.
+        float nonuct_term = this->nonuct_weight * tanhf(child->get_nonuct_score());
+
+        float ucb_value = prior_score + value_score + nonuct_term;
         return ucb_value;
     }
 
diff --git a/core/mcts/ctree/ctree_sampled/lib/cnode.h b/core/mcts/ctree/ctree_sampled/lib/cnode.h
index 7b307d7..be713c8 100644
--- a/core/mcts/ctree/ctree_sampled/lib/cnode.h
+++ b/core/mcts/ctree/ctree_sampled/lib/cnode.h
@@ -12,6 +12,10 @@ namespace tree
     {
         int visit_count, num_children, hidden_state_index_x;
         float reward, pred_value, prior, pred_prob, beta, beta_hat;
+        // NonUCT: eta(theta, a) computed in Python (AsinhGLMSurrogate) and
+        // pushed into this node before selection runs. Defaults to 0 so
+        // nodes never touched by the surrogate behave like standard pUCT.
+        float nonuct_score = 0.0f;
         bool is_root;
         tools::SubTreeValueSet subtree_info;
         std::vector<CNode *> children;                 // shape = (num_children, )
@@ -27,6 +31,12 @@ namespace tree
         float get_qsa(float discount);
         float get_mean_q(float parent_q, float discount);
 
+        // NonUCT: set/get the precomputed asinh-GLM surrogate score for
+        // this node. Called from Python after computing eta(theta, a)
+        // for the candidate action this node represents.
+        void set_nonuct_score(float score) { this->nonuct_score = score; }
+        float get_nonuct_score() { return this->nonuct_score; }
+
         // shape = (agent_num, action_space_size)
         void get_marginal_visit_count(tools::Array2D<int>); // marginal visit_count of sampled actions
         void get_marginal_priors(tools::Array2D<float>);    // marginal node prior of sampled actions
@@ -63,6 +73,10 @@ namespace tree
     public:
         int agent_num, action_space_size, sampled_times, tot_nodes;
         float rho, lam;
+        // NonUCT: weight on the surrogate term in ucb_score. 0.0 = exact
+        // original MAZero behavior (opt-in only via set_nonuct_weight).
+        float nonuct_weight = 0.0f;
+        void set_nonuct_weight(float w) { this->nonuct_weight = w; }
         CNode *node_pool_ptr;
         CNode *root;
         tools::CMinMaxStats minmax_stat; // maintaining qsa for all nodes in the tree except the root.