Nicholaus Halecky added 05_Means_end_Analysis_and__.md  almost 9 years ago

Commit id: 2a39b2ae11ec4c01ebe11f73c1d6f3ab43dac3b6

deletions | additions      

         

# 05 Means-end Analysis and Problem Reduction  Useful for *well-formed* problems.  Lesson preview:  - State Spaces  - Means-End Analysis   - Problems Solving with Means-End Analysis  - Problem Reduction  ## State Spaces  ### The Block Problem  1. You may only move one block at a time.  2. You may only move blocks that have nothing on top of them.  ```  Move(C, Table)  Move(B, C)  Move(A, B)  ```  ### Definition   Have some sort of *Intial State* to a *Goal State*. From state transistions are performed via valid operations.  Q: How do agents know which operations to peform?  ## Means-End Analysis  For each operator that can be applied:  - Apply the current operator to the current state.  - Calculate the difference between the current state and the goal state.  - Prefer the state the minimizes the distance between the new state and goal state.