Probabilistic Programming Bots in Intuitive Physics Game Play

TL;DR

This paper introduces a hybrid probabilistic programming framework for intuitive physics game play, combining physics simulation and neural networks to improve efficiency and performance in environments like Flappy Bird.

Contribution

It presents a novel integration of probabilistic physics simulation with model-free learning to enhance game-playing AI performance.

Findings

The hybrid model outperforms purely model-free approaches.

Combining physics simulation with neural networks improves efficiency.

Empirical results demonstrate success on Flappy Bird.

Abstract

Recent findings suggest that humans deploy cognitive mechanism of physics simulation engines to simulate the physics of objects. We propose a framework for bots to deploy probabilistic programming tools for interacting with intuitive physics environments. The framework employs a physics simulation in a probabilistic way to infer about moves performed by an agent in a setting governed by Newtonian laws of motion. However, methods of probabilistic programs can be slow in such setting due to their need to generate many samples. We complement the model with a model-free approach to aid the sampling procedures in becoming more efficient through learning from experience during game playing. We present an approach where combining model-free approaches (a convolutional neural network in our model) and model-based approaches (probabilistic physics simulation) is able to achieve what neither could alone. This way the model outperforms an all model-free or all model-based approach. We discuss a case study showing empirical results of the performance of the model on the game of Flappy Bird.