Zobrazeno 1 - 10
of 46
pro vyhledávání: '"Nicolas Usunier"'
Does everyone equally benefit from computer vision systems? Answers to this question become more and more important as computer vision systems are deployed at large scale, and can spark major concerns when they exhibit vast performance discrepancies
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e577c2f89c6eeddb848087d297117dfb
http://arxiv.org/abs/2202.07603
http://arxiv.org/abs/2202.07603
Autor:
Nicolas Usunier, Gabriel Synnaeve, Alexander Kirillov, Francisco Massa, Nicolas Carion, Sergey Zagoruyko
Publikováno v:
Computer Vision – ECCV 2020 ISBN: 9783030584511
ECCV (1)
ECCV (1)
We present a new method that views object detection as a direct set prediction problem. Our approach streamlines the detection pipeline, effectively removing the need for many hand-designed components like a non-maximum suppression procedure or ancho
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::375742640347069c0f51433bd286f3aa
https://doi.org/10.1007/978-3-030-58452-8_13
https://doi.org/10.1007/978-3-030-58452-8_13
Autor:
Gabriel Synnaeve, Ronan Collobert, Yossi Adi, Neil Zeghidour, Nicolas Usunier, Vitaliy Liptchinsky
Publikováno v:
ICASSP
Transcribed datasets typically contain speaker identity for each instance in the data. We investigate two ways to incorporate this information during training: Multi-Task Learning and Adversarial Learning. In multi-task learning, the goal is speaker
Publikováno v:
INTERSPEECH
Interspeech 2018
Interspeech 2018, Sep 2018, Hyderabad, India. ⟨10.21437/Interspeech.2018-2414⟩
Interspeech 2018
Interspeech 2018, Sep 2018, Hyderabad, India. ⟨10.21437/Interspeech.2018-2414⟩
State-of-the-art speech recognition systems rely on fixed, hand-crafted features such as mel-filterbanks to preprocess the waveform before the training pipeline. In this paper, we study end-to-end systems trained directly from the raw waveform, build
Autor:
Neil Zeghidour, Nicolas Usunier, Iasonas Kokkinos, Thomas Schatz, Gabriel Synnaeve, Emmanuel Dupoux
Publikováno v:
ICASSP 2018-IEEE International Conference on Acoustics, Speech and Signal Processing
ICASSP 2018-IEEE International Conference on Acoustics, Speech and Signal Processing, Apr 2018, Calgary, Alberta, Canada
HAL
ICASSP 2018-IEEE International Conference on Acoustics, Speech and Signal Processing, Apr 2018, Calgary, Alberta, Canada
HAL
Accepted at ICASSP 2018; International audience; We train a bank of complex filters that operates on the raw waveform and is fed into a convolutional neural network for end-to-end phone recognition. These time-domain filterbanks (TD-filterbanks) are
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::9271025835740daa8dfeea80548ba86b
https://hal.science/hal-01888737/document
https://hal.science/hal-01888737/document
Autor:
Neil Zeghidour, Emmanuel Dupoux, Iasonas Kokkinos, Thomas Schaiz, Nicolas Usunier, Gabriel Synnaeve
Publikováno v:
ICASSP
We train a bank of complex filters that operates on the raw waveform and is fed into a convolutional neural network for end-to-end phone recognition. These time-domain filterbanks (TD-filterbanks) are initialized as an approximation of mel-filterbank
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4b48a0ff973fee2c01e5a8717c7af33c
http://arxiv.org/abs/1711.01161
http://arxiv.org/abs/1711.01161
Publikováno v:
ICCV 2017-International Conference on Computer Vision
ICCV 2017-International Conference on Computer Vision, Sep 2017, Venice, Italy
ICCV
ICCV 2017-International Conference on Computer Vision, Sep 2017, Venice, Italy
ICCV
International audience; In this work we introduce a structured prediction model that endows the Deep Gaussian Conditional Random Field (G-CRF) with a densely connected graph structure. We keep memory and computational complexity under control by expr
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::3716fcc4002b5921b8ff271b8ec29d6c
https://hal.inria.fr/hal-01646293/file/chandra-iccv-2017.pdf
https://hal.inria.fr/hal-01646293/file/chandra-iccv-2017.pdf
Publikováno v:
Machine Learning
Machine Learning, Springer Verlag, 2013, 93 (2-3), pp.227-260. ⟨10.1007/s10994-013-5382-3⟩
Machine Learning, 2013, 93 (2-3), pp.227-260. ⟨10.1007/s10994-013-5382-3⟩
Machine Learning, Springer Verlag, 2013, 93 (2-3), pp.227-260. ⟨10.1007/s10994-013-5382-3⟩
Machine Learning, 2013, 93 (2-3), pp.227-260. ⟨10.1007/s10994-013-5382-3⟩
International audience; Learning to rank is usually reduced to learning to score individual objects, leaving the "ranking" step to a sorting algorithm. In that context, the surrogate loss used for training the scoring function needs to behave well wi
Autor:
Massih-Reza Amini, Nicolas Usunier
This book develops two key machine learning principles: the semi-supervised paradigm and learning with interdependent data. It reveals new applications, primarily web related, that transgress the classical machine learning framework through learning
Publikováno v:
RecSys
International audience; We consider the problem of generating diverse, personalized recommendations such that a small set of recommended items covers a broad range of the user's interests. We represent items in a similarity graph, and we formulate th
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f89b05b6187e06a287d1c1f996d97f3e
https://hal.archives-ouvertes.fr/hal-01387171
https://hal.archives-ouvertes.fr/hal-01387171