Leverage Score Sampling for Complete Mode Coverage in Generative Adversarial Networks

by   Joachim Schreurs, et al.
KU Leuven

Commonly, machine learning models minimize an empirical expectation. As a result, the trained models typically perform well for the majority of the data but the performance may deteriorate on less dense regions of the dataset. This issue also arises in generative modeling. A generative model may overlook underrepresented modes that are less frequent in the empirical data distribution. This problem is known as complete mode coverage. We propose a sampling procedure based on ridge leverage scores which significantly improves mode coverage when compared to standard methods and can easily be combined with any GAN. Ridge Leverage Scores (RLSs) are computed by using an explicit feature map, associated with the next-to-last layer of a GAN discriminator or of a pre-trained network, or by using an implicit feature map corresponding to a Gaussian kernel. Multiple evaluations against recent approaches of complete mode coverage show a clear improvement when using the proposed sampling strategy.


page 7

page 10

page 14

page 16


Comparison of Generative Adversarial Networks Architectures Which Reduce Mode Collapse

Generative Adversarial Networks are known for their high quality outputs...

MGGAN: Solving Mode Collapse using Manifold Guided Training

Mode collapse is a critical problem in training generative adversarial n...

Online Kernel based Generative Adversarial Networks

One of the major breakthroughs in deep learning over the past five years...

Mode Penalty Generative Adversarial Network with adapted Auto-encoder

Generative Adversarial Networks (GAN) are trained to generate sample ima...

Rethinking Generative Coverage: A Pointwise Guaranteed Approach

All generative models have to combat missing modes. The conventional wis...

Rethinking Generative Coverage: A Pointwise Guaranteed Approac

All generative models have to combat missing modes. The conventional wis...

Elastic Interaction Energy-Based Generative Model: Approximation in Feature Space

In this paper, we propose a novel approach to generative modeling using ...

Please sign up or login with your details

Forgot password? Click here to reset