Loading...
Please wait, while we are loading the content...
Similar Documents
Synthesizing Time-Series with Auxiliary Classifier Generative Adversarial Networks
| Content Provider | Semantic Scholar |
|---|---|
| Author | Saeed, Aaqib |
| Copyright Year | 2018 |
| Abstract | In recent years, deep learning has shown state-of-the-art performance on an array of problems such as speech recognition, drug discovery, image segmentation and machine translation. This success is mainly attributed to hand curated datasets either by domain experts or through crowdsourcing. However, other fields (e.g. medicine and assisted living) are lagging behind, where, data curation and sharing are limited due to several factors, for instance, privacy and expensive process of collecting labeled datasets. The motivation of this work is, therefore, to leverage Generative Adversarial Networks (GANs) to develop a framework for generating synthetic time series. This data then could be shared, use to resolve class imbalance and provide better insights into the modeling process. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://www.win.tue.nl/~tozceleb/MScProjects/MS_Thesis_GAN_Timeseries.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
