Play Style Identification Using Low-Level Representations of Play Traces in MicroRTS

TL;DR

This paper introduces an unsupervised CNN-LSTM autoencoder approach to extract meaningful representations from low-level play traces in MicroRTS, enabling better differentiation and exploration of diverse game play styles without domain-specific features.

Contribution

It presents a novel method using unsupervised deep learning to analyze raw play trace data, reducing reliance on handcrafted features and domain knowledge.

Findings

Latent representations effectively separate different play styles.

Approach reduces need for domain-specific feature engineering.

Facilitates exploration of diverse AI play styles.

Abstract

Play style identification can provide valuable game design insights and enable adaptive experiences, with the potential to improve game playing agents. Previous work relies on domain knowledge to construct play trace representations using handcrafted features. More recent approaches incorporate the sequential structure of play traces but still require some level of domain abstraction. In this study, we explore the use of unsupervised CNN-LSTM autoencoder models to obtain latent representations directly from low-level play trace data in MicroRTS. We demonstrate that this approach yields a meaningful separation of different game playing agents in the latent space, reducing reliance on domain expertise and its associated biases. This latent space is then used to guide the exploration of diverse play styles within studied AI players.