update

src/interface.jl

@@ -12,24 +12,24 @@ using ..type, ..mcts
 """ Search the best action to take for a given state and task 
 
 # Arguments
-  - `a::agent`
-    one of Yiem's agents
   - `initial state`
     initial state
-  - `decisionMaker::Function`
+  - `transition::Function`
-    decide what action to take
+    a function that define how the state transitions
-  - `evaluator::Function`
+  - `transitionargs::NamedTuple`
-    assess the value of the state
+    arguments for transition function
-  - `reflector::Function`
+
-    generate lesson from trajectory and reward
+# Keyword Arguments
-  - `isterminal::Function`
+  - `totalsample::Integer`
-    determine whether a given state is a terminal state
+    a number of child state MCTS sample at each node during expansion phase
-  - `n::Integer`
+  - `maxdepth::Integer`
-    how many times action will be sampled from decisionMaker 
+    a number of levels MCTS goes during simulation phase
-  - `w::Float64`
+  - `maxiterations::Integer`
-    exploration weight. Value is usually between 1 to 2. 
+    a number of iteration MCTS goes thru expansion -> simulation -> backpropagation cycle
-    Value 1.0 makes MCTS balance between exploration and exploitation like 50%-50%
+  - `explorationweight::Number`
-    Value 2.0 makes MCTS aggressively search the tree
+    exploration weight controls how much MCTS should explore new state instead of exploit 
+    a known state. 1.0 balance between exploration and exploitation like 50%-50%. 2.0 makes MCTS
+    aggressively explore new state.
 
 # Return
   - `plan::Vector{Dict}`
@@ -41,7 +41,7 @@ julia>
 ```
 
 # TODO
-  [] update docstring
+  [x] update docstring
   [] return best action
 
 # Signature