Zobrazeno 1 - 6
of 6
pro vyhledávání: '"Filho, Telmo Silva"'
Autor:
Filho, Telmo Silva, Song, Hao, Perello-Nieto, Miquel, Santos-Rodriguez, Raul, Kull, Meelis, Flach, Peter
This paper provides both an introduction to and a detailed overview of the principles and practice of classifier calibration. A well-calibrated classifier correctly quantifies the level of uncertainty or confidence associated with its instance-wise p
Externí odkaz:
http://arxiv.org/abs/2112.10327
Autor:
Kull, Meelis, Perello-Nieto, Miquel, Kängsepp, Markus, Filho, Telmo Silva, Song, Hao, Flach, Peter
Class probabilities predicted by most multiclass classifiers are uncalibrated, often tending towards over-confidence. With neural networks, calibration can be improved by temperature scaling, a method to learn a single corrective multiplicative facto
Externí odkaz:
http://arxiv.org/abs/1910.12656
Publikováno v:
AISTATS 2019
Item Response Theory (IRT) aims to assess latent abilities of respondents based on the correctness of their answers in aptitude test items with different difficulty levels. In this paper, we propose the $\beta^3$-IRT model, which models continuous re
Externí odkaz:
http://arxiv.org/abs/1903.04016
Autor:
Moraes, João V.C., Reinaldo, Jéssica T.S., Ferreira-Junior, Manuel, Filho, Telmo Silva, Prudêncio, Ricardo B.C.
Publikováno v:
In Knowledge-Based Systems 15 March 2022 240
Publikováno v:
In Speech Communication February 2022 137:19-34
Autor:
Telles Filho, Telmo Silva
Publikováno v:
Repositório Institucional do Centro Universitário La Salle
Universidade La Salle (UNILASALLE)
instacron:UNILASALLE
Universidade La Salle (UNILASALLE)
instacron:UNILASALLE
Submitted by Leila Silva (leila.silva@unilasalle.edu.br) on 2018-08-01T16:27:45Z No. of bitstreams: 1 Telmo Telles.pdf: 3652754 bytes, checksum: a2769d72323ca65d26f94a167a935ff4 (MD5) Made available in DSpace on 2018-08-01T16:27:45Z (GMT). No. of bit
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od______3056::244656d27fc7361138c113d9f5aed7ce