Conditional generative modeling via learning the latent space
Document Type
Article
Publication Title
arXiv
Abstract
Although deep learning has achieved appealing results on several machine learning tasks, most of the models are deterministic at inference, limiting their application to single-modal settings. We propose a novel general-purpose framework for conditional generation in multimodal spaces, that uses latent variables to model generalizable learning patterns while minimizing a family of regression cost functions. At inference, the latent variables are optimized to find optimal solutions corresponding to multiple output modes. Compared to existing generative solutions, our approach demonstrates faster and stable convergence, and can learn better representations for downstream tasks. Importantly, it provides a simple generic model that can beat highly engineered pipelines tailored using domain expertise on a variety of tasks, while generating diverse outputs. Our codes will be released. Copyright © 2020, The Authors. All rights reserved.
DOI
arXiv:2010.03132
Publication Date
10-7-2020
Keywords
Computer Vision and Pattern Recognition (cs.CV), Machine Learning (cs.LG)
Recommended Citation
S. Ramasinghe, K. Ranasinghe, S. Khan, N. Barnes, and S. Gould, "Conditional Generative Modeling via Learning the Latent Space", 2020, arXiv:2010.03132
Comments
IR Deposit conditions: non-described
Preprint available on arXiv