Zobrazeno 1 - 10
of 38
pro vyhledávání: '"Steven J. Rennie"'
Publikováno v:
EMNLP (1)
BERT-era question answering systems have recently achieved impressive performance on several question-answering (QA) tasks. These systems are based on representations that have been pre-trained on self-supervised tasks such as word masking and senten
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a549899661655e4d92c3c86270483ed4
https://doi.org/10.18653/v1/2020.emnlp-main.87
https://doi.org/10.18653/v1/2020.emnlp-main.87
Autor:
Hui Wu, Xiaoxiao Guo, Steven J. Rennie, Ziad Al-Halah, Yupeng Gao, Kristen Grauman, Rogerio Feris
Publikováno v:
CVPR
Conversational interfaces for the detail-oriented retail fashion domain are more natural, expressive, and user friendly than classical keyword-based search interfaces. In this paper, we introduce the Fashion IQ dataset to support and advance research
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4e022adef28e2e1ca18476c04d2d1e57
http://arxiv.org/abs/1905.12794
http://arxiv.org/abs/1905.12794
Publikováno v:
Acoustical Science and Technology. 37:55-65
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::41182002f6fe77297e9d69b7abcc89e0
https://doi.org/10.1250/ast.37.55
https://doi.org/10.1250/ast.37.55
Autor:
Larry S. Davis, Kristen Grauman, Rogerio Feris, Tushar Nagarajan, Zuxuan Wu, Abhishek Kumar, Steven J. Rennie
Publikováno v:
CVPR
Very deep convolutional neural networks offer excellent recognition results, yet their computational expense limits their impact for many real-world applications. We introduce BlockDrop, an approach that learns to dynamically choose which layers of a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c2f933ae1c010bf0b5a7192368986850
http://arxiv.org/abs/1711.08393
http://arxiv.org/abs/1711.08393
Publikováno v:
AVSP
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::943e2aa40ed27657c7dbfcddc1147409
https://doi.org/10.21437/avsp.2017-13
https://doi.org/10.21437/avsp.2017-13
Publikováno v:
INTERSPEECH
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6559ffd81f27ce15558bec1826e31427
https://doi.org/10.21437/interspeech.2017-852
https://doi.org/10.21437/interspeech.2017-852
Publikováno v:
CVPR
Recently it has been shown that policy-gradient methods for reinforcement learning can be utilized to train deep end-to-end systems directly on non-differentiable metrics for the task at hand. In this paper we consider the problem of optimizing image
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::92178445c3edf1d61d7b4974c648338a
https://doi.org/10.1109/cvpr.2017.131
https://doi.org/10.1109/cvpr.2017.131
Publikováno v:
IEEE Transactions on Audio, Speech, and Language Processing. 21:2244-2254
A variety of first-order methods have recently been proposed for solving matrix optimization problems arising in machine learning. The premise for utilizing such algorithms is that second order information is too expensive to employ, and so simple fi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6390ad58019b1e364fd9d009ca611f98
https://doi.org/10.1109/tasl.2013.2263142
https://doi.org/10.1109/tasl.2013.2263142
Publikováno v:
INTERSPEECH
We describe a collection of acoustic and language modeling techniques that lowered the word error rate of our English conversational telephone LVCSR system to a record 6.6% on the Switchboard subset of the Hub5 2000 evaluation testset. On the acousti
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9b47cf76418cefe869048f1b286bd5a5
https://doi.org/10.21437/interspeech.2016-1460
https://doi.org/10.21437/interspeech.2016-1460
Publikováno v:
ICASSP
In this paper we present a simple and effective method for doing non-linear feature adaptation using Maxout networks. The technique overcomes the need to sample the partition function during training, and overcomes the need to compute the Jacobian te
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::404c031ef83baa93b5efbd78d6c80633
https://doi.org/10.1109/icassp.2016.7472691
https://doi.org/10.1109/icassp.2016.7472691