Abstract
The ability to learn robust, resizable feature representations from unlabeled data has potential applications in a wide variety of machine learning tasks. One way to create such representations is to train deep generative models that can learn to capture the complex distribution of real-world data. Generative adversarial network (GAN) approaches have shown impressive results in producing generative models of images, but relatively little work has been done on evaluating the performance of these methods for the learning representation of natural language, both in supervised and unsupervised settings at the document, sentence, and aspect level. Extensive research validation experiments were performed by leveraging the 20 Newsgroups corpus, the Movie Review (MR) Dataset, and the Finegrained Sentiment Dataset (FSD). Our experimental analysis suggests that GANs can successfully learn representations of natural language texts at all three aforementioned levels.
Highlights
The performance of machine learning (ML) methods is heavily dependent on the choice of data or feature representation to which they are applied
Among the various methods of learning representations (LRs), this paper focuses on deep learning methods: those that are formed by the composition of multiple nonlinear transformations, with the goal of yielding more abstract—and, more useful—representations
We show that a Nash equilibrium [9] under these conditions yields a generator that matches the data distribution
Summary
The performance of machine learning (ML) methods is heavily dependent on the choice of data or feature representation to which they are applied. Much of the actual effort in deploying ML algorithms goes into the design of preprocessing the pipelines and data transformations that result in a representation of the data that can support effective ML. Such feature engineering is important but labor-intensive, which highlights the weakness of current learning algorithms. The emergence of large-scale datasets, such as ImageNet [1], which contains 14,197,122 manually labeled images, has allowed the wider-spread use and popularity of convolutional neural networks (CNNs) even in the unrelated task of medical imaging. In order to expand the scope and ease of applicability of ML, it would be highly desirable to make learning algorithms less dependent on feature engineering, so Algorithms 2018, 11, 164; doi:10.3390/a11100164 www.mdpi.com/journal/algorithms
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
