# Evaluation Function Approximation for Scrabble

## TL;DR

This paper explores learning-based approaches for Scrabble evaluation functions, experimenting with evolutionary algorithms, Bayesian Optimization, and imitation learning to improve move evaluation accuracy.

## Contribution

It introduces a novel approach using imitation learning with neural networks to approximate move rankings in Scrabble, moving beyond traditional simulation-based methods.

## Key findings

- Evolutionary algorithms and Bayesian Optimization were ineffective.
- Supervised imitation learning with neural networks showed promise.
- The approach enables a move ranking system based on raw board input.

## Abstract

The current state-of-the-art Scrabble agents are not learning-based but depend on truncated Monte Carlo simulations and the quality of such agents is contingent upon the time available for running the simulations. This thesis takes steps towards building a learning-based Scrabble agent using self-play. Specifically, we try to find a better function approximation for the static evaluation function used in Scrabble which determines the move goodness at a given board configuration. In this work, we experimented with evolutionary algorithms and Bayesian Optimization to learn the weights for an approximate feature-based evaluation function. However, these optimization methods were not quite effective, which lead us to explore the given problem from an Imitation Learning point of view. We also tried to imitate the ranking of moves produced by the Quackle simulation agent using supervised learning with a neural network function approximator which takes the raw representation of the Scrabble board as the input instead of using only a fixed number of handcrafted features.