@article{Lang_2019, title={Towards Usable Level PCG}, volume={15}, url={https://ojs.aaai.org/index.php/AIIDE/article/view/5247}, DOI={10.1609/aiide.v15i1.5247}, abstractNote={<p>My proposed dissertation work attempts to make procedural content generation (PCG) for game levels easier to use and more expressive for designers. This can be split into three rough layers: improving the expressive range of a dungeon PCG genetic algorithm by using multiobjective optimization, providing a more intuitive way for designers to create fitness variables, and letting them train the PCG system to give them the output they expect. While this work will focus on PCG algorithms for generating game levels, ideally much of the evaluation concept in this paper will be applicable to other forms of PCG. I will hopefully also be able to show that having the designer interact with an expressive PCG system as their ”apprentice” is effective.</p>}, number={1}, journal={Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment}, author={Lang, Eric}, year={2019}, month={Oct.}, pages={213-215} }