Towards Mode Balancing of Generative Models via Diversity Weights
This addresses the need for more diverse outputs in creative applications, though it appears incremental as it builds on existing training schemes.
The paper tackles the problem of low output diversity in generative models by shifting from pure mode coverage to mode balancing, introducing diversity weights to increase diversity, with initial experiments showing potential.
Large data-driven image models are extensively used to support creative and artistic work. Under the currently predominant distribution-fitting paradigm, a dataset is treated as ground truth to be approximated as closely as possible. Yet, many creative applications demand a diverse range of output, and creators often strive to actively diverge from a given data distribution. We argue that an adjustment of modelling objectives, from pure mode coverage towards mode balancing, is necessary to accommodate the goal of higher output diversity. We present diversity weights, a training scheme that increases a model's output diversity by balancing the modes in the training dataset. First experiments in a controlled setting demonstrate the potential of our method. We discuss connections of our approach to diversity, equity, and inclusion in generative machine learning more generally, and computational creativity specifically. An implementation of our algorithm is available at https://github.com/sebastianberns/diversity-weights