update

src/interface.jl

@@ -52,7 +52,8 @@ function runMCTS(
   earlystop::Union{Function,Nothing}=nothing
 )::NamedTuple{(:bestNextState, :bestFinalState),Tuple{T,T}} where {T<:Any}
 
-  root = MCTSNode("root", initialstate, 0, 0, 0, 0, false, nothing, Dict{String,MCTSNode}())
+  root = MCTSNode("root", initialstate, 0, 0, 0, 0, false, nothing, Dict{String,MCTSNode}(),
+                  Dict{Symbol,Any}())
 
   for nth in 1:maxiterations
     node = root
@@ -67,7 +68,7 @@ function runMCTS(
       # do nothing then go directly to backpropagation. It means the end of this iteration
       backpropagate(node, node.reward)
     else
-      expand(node, transition, transitionargs;
+      _ = expand(node, transition, transitionargs;
         totalsample=totalsample)
       leafNode = selectChildNode(node)
       simTrajectoryReward, terminalstate = simulate(leafNode, transition, transitionargs;
@@ -82,6 +83,9 @@ function runMCTS(
       # end
 
       backpropagate(leafNode, simTrajectoryReward)
+
+      # delete all child node, no need for child node that was created during simulation
+      leafNode.children = Dict{String,MCTSNode}() 
     end
 
     # stop if the early stop condition is met

src/mcts.jl

@@ -234,6 +234,7 @@ function expand(node::MCTSNode,transition::Function, transitionargs::NamedTuple;
           totalsample::Integer=3)
 
   nthSample = 0
+  listOfNewNodeId = []
   while true
     nthSample += 1
     if nthSample <= totalsample
@@ -250,12 +251,13 @@ function expand(node::MCTSNode,transition::Function, transitionargs::NamedTuple;
           similar situation to improve decisionMaker and evaluator. 
       """
       if newNodeKey ∉ keys(node.children)
+        push!(listOfNewNodeId, newNodeKey)
         node.children[newNodeKey] = 
           MCTSNode(newNodeKey, newstate, 0, progressvalue, 0, newstate[:reward], 
-            newstate[:isterminal], node, Dict{String, MCTSNode}())
+            newstate[:isterminal], node, Dict{String, MCTSNode}(), Dict{Symbol, Any}())
       end
     else
-      break
+      return listOfNewNodeId
     end
   end
 end
@@ -286,6 +288,7 @@ function simulate(node::MCTSNode, transition::Function, transitionargs::NamedTup
 
   simTrajectoryReward = 0.0
   terminalstate = nothing
+  # listOfSimulatedNodeId = []
   
   for depth in 1:maxdepth
     simTrajectoryReward += node.reward
@@ -293,10 +296,10 @@ function simulate(node::MCTSNode, transition::Function, transitionargs::NamedTup
       terminalstate = node.state
       break
     else
-      expand(node, transition, transitionargs;
-            totalsample=totalsample)
+      _ = expand(node, transition, transitionargs;
+                              totalsample=totalsample)
       node = selectChildNode(node)
-    end
+    end 
   end
 
   return (simTrajectoryReward=simTrajectoryReward, terminalstate=terminalstate)

src/type.jl

@@ -2,6 +2,8 @@ module type
   
 export MCTSNode
 
+using GeneralUtils
+
 # ---------------------------------------------- 100 --------------------------------------------- #
 
 
@@ -49,6 +51,7 @@ mutable struct MCTSNode{T1<:AbstractDict, T2<:AbstractString}
   isterminal::Bool
   parent::Union{MCTSNode, Nothing} 
   children::Dict{String, MCTSNode}
+  etc::Dict{Symbol, Any}  # store anything
 end