Collaborative creativity with Monte-Carlo Tree Search and Convolutional Neural Networks

TL;DR

This paper explores a collaborative drawing system combining Monte Carlo Tree Search with neural networks, revealing that deeper models generate more diverse but often unrecognizable images, highlighting challenges in neural network interpretability.

Contribution

The study integrates Monte Carlo Tree Search with deep convolutional neural networks for collaborative drawing, analyzing how model depth affects image diversity and recognizability.

Findings

Deeper models produce more diverse images.

Humans find images from deep models mostly unrecognizable.

The agent remains highly confident in its objectives despite image quality.

Abstract

We investigate a human-machine collaborative drawing environment in which an autonomous agent sketches images while optionally allowing a user to directly influence the agent's trajectory. We combine Monte Carlo Tree Search with image classifiers and test both shallow models (e.g. multinomial logistic regression) and deep Convolutional Neural Networks (e.g. LeNet, Inception v3). We found that using the shallow model, the agent produces a limited variety of images, which are noticably recogonisable by humans. However, using the deeper models, the agent produces a more diverse range of images, and while the agent remains very confident (99.99%) in having achieved its objective, to humans they mostly resemble unrecognisable 'random' noise. We relate this to recent research which also discovered that 'deep neural networks are easily fooled' \cite{Nguyen2015} and we discuss possible solutions and future directions for the research.