Intriguing properties of neural networks

Intriguing properties of neural networks. Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. arXiv:1312.6199 [cs], February, 2014. arXiv: 1312.6199

Paper abstract bibtex

Deep neural networks are highly expressive models that have recently achieved state of the art performance on speech and visual recognition tasks. While their expressiveness is the reason they succeed, it also causes them to learn uninterpretable solutions that could have counter-intuitive properties. In this paper we report two such properties.

@article{szegedy_intriguing_2014,
	title = {Intriguing properties of neural networks},
	url = {http://arxiv.org/abs/1312.6199},
	abstract = {Deep neural networks are highly expressive models that have recently achieved state of the art performance on speech and visual recognition tasks. While their expressiveness is the reason they succeed, it also causes them to learn uninterpretable solutions that could have counter-intuitive properties. In this paper we report two such properties.},
	language = {en},
	urldate = {2020-07-24},
	journal = {arXiv:1312.6199 [cs]},
	author = {Szegedy, Christian and Zaremba, Wojciech and Sutskever, Ilya and Bruna, Joan and Erhan, Dumitru and Goodfellow, Ian and Fergus, Rob},
	month = feb,
	year = {2014},
	note = {arXiv: 1312.6199},
	keywords = {⛔ No DOI found},
}

Downloads: 0

{"_id":"i2pp8HTT5qaAxaGFZ","bibbaseid":"szegedy-zaremba-sutskever-bruna-erhan-goodfellow-fergus-intriguingpropertiesofneuralnetworks-2014","author_short":["Szegedy, C.","Zaremba, W.","Sutskever, I.","Bruna, J.","Erhan, D.","Goodfellow, I.","Fergus, R."],"bibdata":{"bibtype":"article","type":"article","title":"Intriguing properties of neural networks","url":"http://arxiv.org/abs/1312.6199","abstract":"Deep neural networks are highly expressive models that have recently achieved state of the art performance on speech and visual recognition tasks. While their expressiveness is the reason they succeed, it also causes them to learn uninterpretable solutions that could have counter-intuitive properties. In this paper we report two such properties.","language":"en","urldate":"2020-07-24","journal":"arXiv:1312.6199 [cs]","author":[{"propositions":[],"lastnames":["Szegedy"],"firstnames":["Christian"],"suffixes":[]},{"propositions":[],"lastnames":["Zaremba"],"firstnames":["Wojciech"],"suffixes":[]},{"propositions":[],"lastnames":["Sutskever"],"firstnames":["Ilya"],"suffixes":[]},{"propositions":[],"lastnames":["Bruna"],"firstnames":["Joan"],"suffixes":[]},{"propositions":[],"lastnames":["Erhan"],"firstnames":["Dumitru"],"suffixes":[]},{"propositions":[],"lastnames":["Goodfellow"],"firstnames":["Ian"],"suffixes":[]},{"propositions":[],"lastnames":["Fergus"],"firstnames":["Rob"],"suffixes":[]}],"month":"February","year":"2014","note":"arXiv: 1312.6199","keywords":"⛔ No DOI found","bibtex":"@article{szegedy_intriguing_2014,\n\ttitle = {Intriguing properties of neural networks},\n\turl = {http://arxiv.org/abs/1312.6199},\n\tabstract = {Deep neural networks are highly expressive models that have recently achieved state of the art performance on speech and visual recognition tasks. While their expressiveness is the reason they succeed, it also causes them to learn uninterpretable solutions that could have counter-intuitive properties. In this paper we report two such properties.},\n\tlanguage = {en},\n\turldate = {2020-07-24},\n\tjournal = {arXiv:1312.6199 [cs]},\n\tauthor = {Szegedy, Christian and Zaremba, Wojciech and Sutskever, Ilya and Bruna, Joan and Erhan, Dumitru and Goodfellow, Ian and Fergus, Rob},\n\tmonth = feb,\n\tyear = {2014},\n\tnote = {arXiv: 1312.6199},\n\tkeywords = {⛔ No DOI found},\n}\n\n\n\n","author_short":["Szegedy, C.","Zaremba, W.","Sutskever, I.","Bruna, J.","Erhan, D.","Goodfellow, I.","Fergus, R."],"key":"szegedy_intriguing_2014","id":"szegedy_intriguing_2014","bibbaseid":"szegedy-zaremba-sutskever-bruna-erhan-goodfellow-fergus-intriguingpropertiesofneuralnetworks-2014","role":"author","urls":{"Paper":"http://arxiv.org/abs/1312.6199"},"keyword":["⛔ No DOI found"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/howcanoewang","dataSources":["CKyz8yvoYYbDNyw6n","aXmRAq63YsH7a3ufx","Wsv2bQ4jPuc7qme8R","z5vjAHmuibbBBnvdc"],"keywords":["⛔ no doi found"],"search_terms":["intriguing","properties","neural","networks","szegedy","zaremba","sutskever","bruna","erhan","goodfellow","fergus"],"title":"Intriguing properties of neural networks","year":2014}