Compressive Learning of Generative Networks
This addresses the problem of high computational costs for researchers and practitioners training generative models on large datasets, representing an incremental improvement.
The paper tackles the computational expense of training generative networks on large-scale datasets by compressing datasets into a single sketch vector and optimizing a cost function that approximates Maximum Mean Discrepancy, achieving time- and memory-efficient training.
Generative networks implicitly approximate complex densities from their sampling with impressive accuracy. However, because of the enormous scale of modern datasets, this training process is often computationally expensive. We cast generative network training into the recent framework of compressive learning: we reduce the computational burden of large-scale datasets by first harshly compressing them in a single pass as a single sketch vector. We then propose a cost function, which approximates the Maximum Mean Discrepancy metric, but requires only this sketch, which makes it time- and memory-efficient to optimize.