65 citations · 152 across the 20 of their papers we have counts for
12 papers · 1 filter
Understanding Latent Diffusability via Fisher Geometry
Jing Gu, Morteza Mardani, Wonjun Lee +2
Diffusion models often degrade in latent spaces, yet the formal causes remain poorly understood. We quantify latent-space diffusability via the rate of change of the Minimum Mean S…
Stein Discrepancy for Unsupervised Domain Adaptation
Anneke von Seeger, Dongmian Zou, Gilad Lerman
Unsupervised domain adaptation (UDA) aims to improve model performance on an unlabeled target domain using a related, labeled source domain. A common approach aligns source and tar…
Improving Hyperbolic Representations via Gromov-Wasserstein Regularization
Yifei Yang, Wonjun Lee, Dongmian Zou +1
Hyperbolic representations have shown remarkable efficacy in modeling inherent hierarchies and complexities within data structures. Hyperbolic neural networks have been commonly ap…
Monotone Generative Modeling via a Gromov-Monge Embedding
Wonjun Lee, Yifei Yang, Dongmian Zou +1
Generative adversarial networks (GANs) are popular for generative tasks; however, they often require careful architecture selection, extensive empirical tuning, and are prone to mo…
Hyperbolic Convolution via Kernel Point Aggregation
Eric Qu, Dongmian Zou
Learning representations according to the underlying geometry is of vital importance for non-Euclidean data. Studies have revealed that the hyperbolic space can effectively embed h…
An Unpooling Layer for Graph Generation
Yinglong Guo, Dongmian Zou, Gilad Lerman
We propose a novel and trainable graph unpooling layer for effective graph generation. Given a graph with features, the unpooling layer enlarges this graph and learns its desired n…
Robust Vector Quantized-Variational Autoencoder
Chieh-Hsin Lai, Dongmian Zou, Gilad Lerman
Image generative models can learn the distributions of the training data and consequently generate examples by sampling from these distributions. However, when the training dataset…
Autoencoding Hyperbolic Representation for Adversarial Generation
Eric Qu, Dongmian Zou
With the recent advance of geometric deep learning, neural networks have been extensively used for data in non-Euclidean domains. In particular, hyperbolic neural networks have pro…
Novelty Detection via Robust Variational Autoencoding
Chieh-Hsin Lai, Dongmian Zou, Gilad Lerman
We propose a new method for novelty detection that can tolerate high corruption of the training points, whereas previous works assumed either no or very low corruption. Our method…
Robust Subspace Recovery Layer for Unsupervised Anomaly Detection
Chieh-Hsin Lai, Dongmian Zou, Gilad Lerman
We propose a neural network for unsupervised anomaly detection with a novel robust subspace recovery layer (RSR layer). This layer seeks to extract the underlying subspace from a l…
Encoding Robust Representation for Graph Generation
Dongmian Zou, Gilad Lerman
Generative networks have made it possible to generate meaningful signals such as images and texts from simple noise. Recently, generative methods based on GAN and VAE were develope…
Lipschitz Properties for Deep Convolutional Networks
Radu Balan, Maneesh Singh, Dongmian Zou
In this paper we discuss the stability properties of convolutional neural networks. Convolutional neural networks are widely used in machine learning. In classification they are ma…