Maren Aiwszus received the AIIDE 2020 Best Student Paper Award for the paper TOAD-GAN: Coherent Style Level Generation from a Single Example, which she presented at the Sixteenth International AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE). The conference took place online from 19 until 23 October 2020. Maren Aiwszus wrote the paper together with Frederik Schubert and Prof. Bodo Rosenhahn from L3S and the Institute of Information Processing (LUH).
TOAD-GAN (Token-based One-shot Arbitrary Dimension Generative Adversarial Network) is a novel Procedural Content Generation (PCG) algorithm that generates token-based video game levels.
Click here for Maren Aiwszus´ video presentation of TOAD-GAN on Youtube:
The Code was released on Github. With the Demonstrator, you can play the generated levels.