Computational creativity

Abstract: In the space of only a few years, deep generative modeling has revolutionized how we think of artificial creativity, yielding autonomous systems which produce original images, music, and text. Inspired by these successes, researchers are now applying deep generative modeling techniques to the generation and optimization of molecules—in our review we found 45 papers on the subject published in the past two years. These works point to a future where such systems will be used to generate lead molecules, greatly reducing resources spent downstream synthesizing and characterizing bad leads in the lab. In this review we survey the increasingly complex landscape of models and representation schemes that have been proposed. The four classes of techniques we describe are recursive neural networks, autoencoders, generative adversarial networks, and reinforcement learning. After first discussing some of the mathematical fundamentals of each technique, we draw high level connections and comparisons with other techniques and expose the pros and cons of each. Several important high level themes emerge as a result of this work, including the shift away from the SMILES string representation of molecules towards more sophisticated representations such as graph grammars and 3D representations, the importance of reward function design, the need for better standards for benchmarking and testing, and the benefits of adversarial training and reinforcement learning over maximum likelihood based training.

Abstract: Notions relating to computational systems exhibiting creative behaviours have been explored since the very early days of computer science, and the field of Computational Creativity research has formed in the last dozen years to scientifically explore the potential of such systems. We describe this field via a working definition; a brief history of seminal work; an exploration of the main issues, technologies and ideas; and a look towards future directions. As a society, we are jealous of our creativity: creative people and their contributions to cultural progression are highly valued. Moreover, creative behaviour in people draws on a full set of intelligent abilities, so simulating such behaviour represents a serious technical challenge for Artificial Intelligence research. As such, we believe it is fair to characterise Computational Creativity as a frontier for AI research beyond all others—maybe, even, the final frontier.

Abstract: Even in the absence of external reward, babies and scientists and others explore their world. Using some sort of adaptive predictive world model, they improve their ability to answer questions such as what happens if I do this or that? They lose interest in both the predictable things and those predicted to remain unpredictable despite some effort. One can design curious robots that do the same. The author’s basic idea (1990, 1991) for doing so is a reinforcement learning (RL) controller is rewarded for action sequences that improve the predictor. Here, this idea is revisited in the context of recent results on optimal predictors and optimal RL machines. Several new variants of the basic principle are proposed. Finally, it is pointed out how the fine arts can be formally understood as a consequence of the principle: given some subjective observer, great works of art and music yield observation histories exhibiting more novel, previously unknown compressibility/regularity/predictability (with respect to th...

Abstract: I summarise and attempt to clarify some concepts presented in and arising from Margaret Boden's (1990) descriptive hierarchy of creativity, by beginning to formalise the ideas she proposes. The aim is to move towards a model which allows detailed comparison, and hence better understanding, of systems which exhibit behaviour which would be called ''creative'' in humans. The work paves the way for the description of naturalistic, multi-agent creative AI systems, which create in a societal context. I demonstrate some simple reasoning about creative behaviour based on the new framework, to show how it might be useful for the analysis and study of creative systems. In particular, I identify some crucial properties of creative systems, in terms of the framework components, some of which may usefully be proven a priori of a given system. I suggest that Boden's descriptive framework, once elaborated in detail, is more uniform and more powerful than it first appears.