Liangliang Shi

Shanghai Jiao Tong University | SJTU

Optimal transport (OT) is attracting increasing attention in machine learning. It aims to transport a source distribution to a target one at minimal cost. In its vanilla form, the source and target distributions are predetermined, which contracts to the real-world case involving undetermined targets. In this paper, we propose Doubly Bounded Optimal...

For Generative Adversarial Networks which map a latent distribution to the target distribution, in this paper, we study how the sampling in latent space can affect the generation performance, especially for images. We observe that, as the neu-ral generator is a continuous function, two close samples in latent space would be mapped into two nearby i...

Despite its success, generative adversarial networks (GANs) still suffer from mode collapse, i.e., the generator can only map latent variables to a partial set of modes in the target distribution. In this paper, we analyze and seek to regularize this issue with an independent and identically distributed (IID) sampling perspective and emphasize that...

Despite its success, generative adversarial networks (GANs) still suffer from mode collapse, namely the generator can only map latent variables to a partial set of modes of the target distribution. In this paper, we analyze and try to regularize this issue with an independent and identically distributed (IID) sampling perspective and emphasize that...

This paper proposes new ways of sample mixing by thinking of the process as generation of barycenter in a metric space for data augmentation. First, we present an optimal-transport-based mixup technique to generate Wasserstein barycenter which works well on images with clean background and is empirically shown complementary to existing mixup method...

Causality learning has been an important tool for decision making, especially for financial analytics. Given the time series data, most existing works construct the causality network with the traditional regression models and estimate the causality by pairs. To fulfil a holistic one-shot inference procedure over the whole network, we propose a new...

Point process is an expressive tool in learning temporal event sequence which is ubiquitous in real-world applications. Traditional predictive models are based on maximum likelihood estimation (MLE). This paper aims to improve MLE by discriminative and adversarial learning. The initial model is learned by MLE explaining the joint distribution of th...