Build a Game-Playing Agent

Is adversarial search correctly implemented using iterative deepening, minimax, and alpha-beta pruning?

The minimax and alphabeta functions pass all test cases.

Submission must contain source code, heuristic report, and research summary report.

All required file included.

Have at least three (3) evaluation heuristics besides null_score(), open_move_score(), and improved_score() been implemented and analyzed?

At least three evaluation functions are implemented and analyzed.

Has the performance of agents against the testing agents been adequately described?

A brief report lists (using a table and any appropriate visualizations) and verbally describes the performance of agents using the implemented evaluation functions. Performance data includes results from tournament.py comparing (at a minimum) the best performing student heuristic against the ID_Improved agent.

Does the report make a recommendation about the best evaluation function, and is this recommendation adequately justified?

The report makes a recommendation about which evaluation function should be used and justifies the recommendation with at least three reasons supported by the data.

Completeness

The write up is approximately 1 page (500 words) and includes a summary of the paper (including new techniques introduced), and the key results (if any) that were achieved.