Creative divergent synthesis with generative models
It addresses the challenge of enabling generative models to exhibit creative divergence for applications in domains like image, audio, or video generation, but it is incremental as it offers perspectives and preliminary results.
The paper tackles the problem that generative models typically model data distributions without extrapolating, limiting creativity, and proposes a novel training objective called Bounded Adversarial Divergence (BAD) with preliminary results.
Machine learning approaches now achieve impressive generation capabilities in numerous domains such as image, audio or video. However, most training \& evaluation frameworks revolve around the idea of strictly modelling the original data distribution rather than trying to extrapolate from it. This precludes the ability of such models to diverge from the original distribution and, hence, exhibit some creative traits. In this paper, we propose various perspectives on how this complicated goal could ever be achieved, and provide preliminary results on our novel training objective called \textit{Bounded Adversarial Divergence} (BAD).