Adversarial network training using higher-order moments in a modified Wasserstein distance
Generative-adversarial networks (GANs) have been used to produce data closely resembling example data in a compressed, latent space that is close to sufficient for reconstruction in the original vector space. The Wasserstein metric has been used as an alternative to binary cross-entropy, producing more numerically stable GANs with greater mode covering behavior. Here, a generalization of the Wasserstein distance, using higher-order moments than the mean, is derived. Training a GAN with this higher-order Wasserstein metric is demonstrated to exhibit superior performance, even when adjusted for slightly higher computational cost. This is illustrated generating synthetic antibody sequences.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
On Higher-order Moments in Adam
In this paper, we investigate the popular deep learning optimization routine, Adam, from the perspective of statistical moments. While Adam is an adaptive lower-order moment based (of the stochastic gradient) method, we …
An investigation of higher order moments of empirical financial data and the implications to risk
Here, we analyse the behaviour of the higher order standardised moments of financial time series when we truncate a large data set into smaller and smaller subsets, referred to below as time windows. We look at the effec…
Time SeriesTime Series AnalysisNon-stationary GARCH modelling for fitting higher order moments of financial series within moving time windows
Here, we have analysed a GARCH(1,1) model with the aim to fit higher order moments for different companies' stock prices. When we assume a gaussian conditional distribution, we fail to capture any empirical data when fit…
Time SeriesTime Series AnalysisAccelerated and Improved Stabilization for High Order Moments of Racah Polynomials
One of the most effective orthogonal moments, discrete Racah polynomials (DRPs) and their moments are used in many disciplines of sciences, including image processing, and computer vision. Moments are the projections of …
Vocal Bursts Intensity PredictionEfficient Computation of Higher Order 2D Image Moments using the Discrete Radon Transform
Geometric moments and moment invariants of image artifacts have many uses in computer vision applications, e.g. shape classification or object position and orientation. Higher order moments are of interest to provide add…
Position