JSON-Bag: A generic game trajectory representation

TL;DR

JSON-Bag is a novel, generic method for representing game trajectories using tokenized JSON descriptions and Jensen-Shannon distance, enabling effective classification and feature extraction across various tabletop games.

Contribution

The paper introduces JSON-Bag, a new approach for representing game trajectories that improves classification accuracy and sample efficiency over hand-crafted features.

Findings

JSON-Bag outperforms hand-crafted features in most classification tasks.

Using JSON-Bag prototypes enhances sample efficiency in N-shot classification.

JSD between JSON-Bag prototypes correlates with policy distances across games.

Abstract

We introduce JSON Bag-of-Tokens model (JSON-Bag) as a method to generically represent game trajectories by tokenizing their JSON descriptions and apply Jensen-Shannon distance (JSD) as distance metric for them. Using a prototype-based nearest-neighbor search (P-NNS), we evaluate the validity of JSON-Bag with JSD on six tabletop games: 7 Wonders, Dominion, Sea Salt and Paper, Can't Stop, Connect4, Dots and boxes; each over three game trajectory classification tasks: classifying the playing agents, game parameters, or game seeds that were used to generate the trajectories. Our approach outperforms a baseline using hand-crafted features in the majority of tasks. Evaluating on N-shot classification suggests using JSON-Bag prototype to represent game trajectory classes is also sample efficient. Additionally, we demonstrate JSON-Bag ability for automatic feature extraction by treating tokens as individual features to be used in Random Forest to solve the tasks above, which significantly improves accuracy on underperforming tasks. Finally, we show that, across all six games, the JSD between JSON-Bag prototypes of agent classes highly correlates with the distances between agents' policies.