KernelNet: A Data-Dependent Kernel Parameterization for Deep Generative Modeling

12/02/2019
by   Yufan Zhou, et al.
24

Learning with kernels is an often resorted tool in modern machine learning. Standard approaches for this type of learning use a predefined kernel that requires careful selection of hyperparameters. To mitigate this burden, we propose in this paper a framework to construct and learn a data-dependent kernel based on random features and implicit spectral distributions (Fourier transform of the kernel) parameterized by deep neural networks. We call the constructed network KernelNet, and apply it for deep generative modeling in various scenarios, including variants of the MMD-GAN and an implicit Variational Autoencoder (VAE), the two popular learning paradigms in deep generative models. Extensive experiments show the advantages of the proposed KernelNet, consistently achieving better performance compared to related methods.

READ FULL TEXT

page 10

page 12

page 14

page 23

research
02/26/2019

Implicit Kernel Learning

Kernels are powerful and versatile tools in machine learning and statist...
research
06/02/2017

On Unifying Deep Generative Models

Deep generative models have achieved impressive success in recent years....
research
02/22/2017

Adversarial examples for generative models

We explore methods of producing adversarial examples on deep generative ...
research
09/20/2020

Factorized Deep Generative Models for Trajectory Generation with Spatiotemporal-Validity Constraints

Trajectory data generation is an important domain that characterizes the...
research
06/15/2022

Training Discrete Deep Generative Models via Gapped Straight-Through Estimator

While deep generative models have succeeded in image processing, natural...
research
01/24/2019

Deep Generative Learning via Variational Gradient Flow

We propose a general framework to learn deep generative models via Varia...
research
11/16/2018

Deep Knockoffs

This paper introduces a machine for sampling approximate model-X knockof...

Please sign up or login with your details

Forgot password? Click here to reset