Generating Multi-Categorical Samples with Generative Adversarial Networks. Camino, R., Hammerschmidt, C., & State, R. July, 2018. _eprint: 1807.01202![Generating Multi-Categorical Samples with Generative Adversarial Networks [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex We propose a method to train generative adversarial networks on mutivariate feature vectors representing multiple categorical values. In contrast to the continuous domain, where GAN-based methods have delivered considerable results, GANs struggle to perform equally well on discrete data. We propose and compare several architectures based on multiple (Gumbel) softmax output layers taking into account the structure of the data. We evaluate the performance of our architecture on datasets with different sparsity, number of features, ranges of categorical values, and dependencies among the features. Our proposed architecture and method outperforms existing models.
@article{camino_generating_2018,
title = {Generating {Multi}-{Categorical} {Samples} with {Generative} {Adversarial} {Networks}},
url = {http://arxiv.org/abs/1807.01202},
abstract = {We propose a method to train generative adversarial networks on mutivariate feature vectors representing multiple categorical values. In contrast to the continuous domain, where GAN-based methods have delivered considerable results, GANs struggle to perform equally well on discrete data. We propose and compare several architectures based on multiple (Gumbel) softmax output layers taking into account the structure of the data. We evaluate the performance of our architecture on datasets with different sparsity, number of features, ranges of categorical values, and dependencies among the features. Our proposed architecture and method outperforms existing models.},
author = {Camino, Ramiro and Hammerschmidt, Christian and State, Radu},
month = jul,
year = {2018},
note = {\_eprint: 1807.01202},
}
Downloads: 0
{"_id":"4semkKRgTzGK8ugHw","bibbaseid":"camino-hammerschmidt-state-generatingmulticategoricalsampleswithgenerativeadversarialnetworks-2018","authorIDs":[],"author_short":["Camino, R.","Hammerschmidt, C.","State, R."],"bibdata":{"bibtype":"article","type":"article","title":"Generating Multi-Categorical Samples with Generative Adversarial Networks","url":"http://arxiv.org/abs/1807.01202","abstract":"We propose a method to train generative adversarial networks on mutivariate feature vectors representing multiple categorical values. In contrast to the continuous domain, where GAN-based methods have delivered considerable results, GANs struggle to perform equally well on discrete data. We propose and compare several architectures based on multiple (Gumbel) softmax output layers taking into account the structure of the data. We evaluate the performance of our architecture on datasets with different sparsity, number of features, ranges of categorical values, and dependencies among the features. Our proposed architecture and method outperforms existing models.","author":[{"propositions":[],"lastnames":["Camino"],"firstnames":["Ramiro"],"suffixes":[]},{"propositions":[],"lastnames":["Hammerschmidt"],"firstnames":["Christian"],"suffixes":[]},{"propositions":[],"lastnames":["State"],"firstnames":["Radu"],"suffixes":[]}],"month":"July","year":"2018","note":"_eprint: 1807.01202","bibtex":"@article{camino_generating_2018,\n\ttitle = {Generating {Multi}-{Categorical} {Samples} with {Generative} {Adversarial} {Networks}},\n\turl = {http://arxiv.org/abs/1807.01202},\n\tabstract = {We propose a method to train generative adversarial networks on mutivariate feature vectors representing multiple categorical values. In contrast to the continuous domain, where GAN-based methods have delivered considerable results, GANs struggle to perform equally well on discrete data. We propose and compare several architectures based on multiple (Gumbel) softmax output layers taking into account the structure of the data. We evaluate the performance of our architecture on datasets with different sparsity, number of features, ranges of categorical values, and dependencies among the features. Our proposed architecture and method outperforms existing models.},\n\tauthor = {Camino, Ramiro and Hammerschmidt, Christian and State, Radu},\n\tmonth = jul,\n\tyear = {2018},\n\tnote = {\\_eprint: 1807.01202},\n}\n\n","author_short":["Camino, R.","Hammerschmidt, C.","State, R."],"key":"camino_generating_2018","id":"camino_generating_2018","bibbaseid":"camino-hammerschmidt-state-generatingmulticategoricalsampleswithgenerativeadversarialnetworks-2018","role":"author","urls":{"Paper":"http://arxiv.org/abs/1807.01202"},"downloads":0},"bibtype":"article","biburl":"https://api.zotero.org/users/3522498/collections/AW3NX4WW/items?key=kdJ5QIjIIc7oy1mYjjz70Rv2&format=bibtex&limit=100","creationDate":"2021-03-14T11:19:06.390Z","downloads":0,"keywords":[],"search_terms":["generating","multi","categorical","samples","generative","adversarial","networks","camino","hammerschmidt","state"],"title":"Generating Multi-Categorical Samples with Generative Adversarial Networks","year":2018,"dataSources":["dwrmKCbrccWf5bf2H"]}