Active Divergence with Generative Deep Learning -- A Survey and Taxonomy

TL;DR

This paper surveys methods for actively diverging generative deep learning models from their training data to enhance computational creativity, providing a taxonomy and highlighting future research directions.

Contribution

It introduces a comprehensive taxonomy of active divergence techniques and reviews current approaches to enable creative applications of generative models.

Findings

Taxonomy of active divergence methods

Identification of key challenges and opportunities

Guidelines for future research in creative AI

Abstract

Generative deep learning systems offer powerful tools for artefact generation, given their ability to model distributions of data and generate high-fidelity results. In the context of computational creativity, however, a major shortcoming is that they are unable to explicitly diverge from the training data in creative ways and are limited to fitting the target data distribution. To address these limitations, there have been a growing number of approaches for optimising, hacking and rewriting these models in order to actively diverge from the training data. We present a taxonomy and comprehensive survey of the state of the art of active divergence techniques, highlighting the potential for computational creativity researchers to advance these methods and use deep generative models in truly creative systems.