Imitation learning has been demonstrated as a useful technique in automatic game testing and the development of believable Non-Player Characters (NPCs). However, imitation learning methods typically focus on learning a policy to complete a task without consideration about the playstyle used. In this work we consider the case where the task is to imitate a given playstyle. We defined a player's playstyle based on the strategies they use in order to complete the overall task. This has been achieved by rewarding a learning agent based on the similarity of the agent and demonstration trajectories, within a learnt representation space. This allows the playstyle to be learnt in levels that differ to the one the demonstrations were collected in.

The automated evaluation of creative products promises both good-and-scalable creativity assessments and new forms of visual analysis of whole corpora. Where creative works are not ‘born digital’, such automated evaluation requires fast and frugal ways of transforming them into data representations that can be meaningfully assessed with common creativity metrics like novelty. In this paper, we report the results of training a Spatiotemporal DeepInfomax Variational Autoencoder (STDIM-VAE) on a digital photo pool of 162 LEGO ducks to generate a phenotypical landscape of clusters of similar ducks and dissimilarity scores for individual ducks. Visual inspection suggests that our system produces plausible results from image pixels alone. We conclude that under certain conditions, STDIM-VAEs may provide fast and frugal ways of automatically assessing corpora of creative works.

Player clustering when applied to the field of video games has several potential applications. For example, the evaluation of the composition of a player base or the generation of AI agents with identified playing styles. These agents can then be used for either the testing of new game content or used directly to enhance a player’s gaming experience. Most current player clustering techniques focus on the use of internal game variables. This raises two main issues: (1) the availability of game variables, as source code access is required to log them and hence limits the data sources that can be used, and (2) the choice of game variables can introduce unintended bias in the types of play style extracted. In this work, a hybrid unsupervised frame encoder and a ‘reference-based’ clustering algorithm are both proposed and combined to allow clustering from raw game play videos. It is shown that the proposed methods are most beneficial when the types of play styles are unknown.