Zobrazeno 1 - 10
of 142
pro vyhledávání: '"Alexandros Potamianos"'
Autor:
Athanasia Zlatintsi, Petros Koutras, Georgios Evangelopoulos, Nikolaos Malandrakis, Niki Efthymiou, Katerina Pastra, Alexandros Potamianos, Petros Maragos
Publikováno v:
EURASIP Journal on Image and Video Processing, Vol 2017, Iss 1, Pp 1-24 (2017)
Abstract Research related to computational modeling for machine-based understanding requires ground truth data for training, content analysis, and evaluation. In this paper, we present a multimodal video database, namely COGNIMUSE, annotated with sen
Externí odkaz:
https://doaj.org/article/8227720ba6b24450ae51c338d8c28908
Publikováno v:
Interspeech 2021.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d2fc52735a9bbc71ff21f45be64616ba
https://doi.org/10.21437/interspeech.2021-1739
https://doi.org/10.21437/interspeech.2021-1739
Publikováno v:
NAACL-HLT
In this work we explore Unsupervised Domain Adaptation (UDA) of pretrained language models for downstream tasks. We introduce UDALM, a fine-tuning procedure, using a mixed classification and Masked Language Model loss, that can adapt to the target do
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1e2e7c2c8ea5a5f874228e9cfe3f08b4
https://doi.org/10.18653/v1/2021.naacl-main.203
https://doi.org/10.18653/v1/2021.naacl-main.203
Publikováno v:
INTERSPEECH
In this work we propose a machine learning model for depression detection from transcribed clinical interviews. Depression is a mental disorder that impacts not only the subject's mood but also the use of language. To this end we use a Hierarchical A
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::12564fdbbb9d9b6e39df02e63a3e118c
https://doi.org/10.21437/interspeech.2020-2819
https://doi.org/10.21437/interspeech.2020-2819
Publikováno v:
INTERSPEECH
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::418a4e2c8a1b656feee0ea2824a38e4a
https://doi.org/10.21437/interspeech.2019-3243
https://doi.org/10.21437/interspeech.2019-3243
Autor:
Theodoros Giannakopoulos, Aggelina Chatziagapi, Shrikanth S. Narayanan, Alexandros Potamianos, Athanasios Katsamanis, Georgios Pantazopoulos, Spiros Dimopoulos, Dimitris Sgouropoulos
Publikováno v:
CBMI
This paper demonstrates the utilization of Oliver 11https://behavioralsignals.com/oliver/, the speech emotion recognition (SER) API created by Behavioral Signals, in the context of a movie content visualization application. Oliver API provides an emo
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0445f15cf58e0b2d8918a288aa71d88a
https://doi.org/10.1109/cbmi.2019.8877398
https://doi.org/10.1109/cbmi.2019.8877398
Publikováno v:
ACL (1)
In this paper, we present a novel approach for incorporating external knowledge in Recurrent Neural Networks (RNNs). We propose the integration of lexicon features into the self-attention mechanism of RNN-based architectures. This form of conditionin
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::65b0f0e256b47ff22f7b42b2424f68b7
http://arxiv.org/abs/1906.03674
http://arxiv.org/abs/1906.03674
Publikováno v:
NAACL-HLT (1)
In traditional Distributional Semantic Models (DSMs) the multiple senses of a polysemous word are conflated into a single vector space representation. In this work, we propose a DSM that learns multiple distributional representations of a word based
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9ed368401801439a5eed16e1b9931e75
https://doi.org/10.18653/v1/n19-1110
https://doi.org/10.18653/v1/n19-1110
Publikováno v:
NAACL-HLT (1)
A growing number of state-of-the-art transfer learning methods employ language models pretrained on large generic corpora. In this paper we present a conceptually simple and effective transfer learning approach that addresses the problem of catastrop
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7b4bad7f75cf8d9d2bf70172b4c63240
https://doi.org/10.18653/v1/n19-1213
https://doi.org/10.18653/v1/n19-1213
Publikováno v:
NAACL-HLT (1)
Neural sequence-to-sequence models are currently the dominant approach in several natural language processing tasks, but require large parallel corpora. We present a sequence-to-sequence-to-sequence autoencoder (SEQˆ3), consisting of two chained enc
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::52d2001f218b62f33d948c34f5337947
https://doi.org/10.18653/v1/n19-1071
https://doi.org/10.18653/v1/n19-1071