Constrained bayesian active learning of linear classifier

[en] In this paper, an on-line interactive method is proposed for learning a linear classifier. This problem is studied within the Active Learning (AL) framework where the learning algorithm sequentially chooses unlabelled training samples and requests their class labels from an oracle in order to learn the classifier with the least queries to the oracle possible. Additionally' a constraint is introduced into this interactive learning process which limits the percentage of the samples from one “unwanted” class under a certain threshold. An optimal AL solution is derived and implemented with a sophisticated, accurate and fast Bayesian Learning method, the Expectation Propagation (EP) and its performance is demonstrated through numerical simulations.

Disciplines :

Computer science

Author, co-author :

Tsakmalis, Anestis ; University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT)

Chatzinotas, Symeon ; University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT)

Ottersten, Björn ; University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT)

External co-authors :

yes

Language :

English

Title :

Constrained bayesian active learning of linear classifier

Publication date :

2018

Event name :

2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Event place :

Calgary, Canada

Event date :

15-04-2018 to 20-04-2018

Audience :

International

Main work title :

2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Publisher :

IEEE

ISBN/EAN :

978-1-5386-4658-8

Peer reviewed :

Peer reviewed

Focus Area :

Computational Sciences

European Projects :

H2020 - 742648 - AGNOSTIC - Actively Enhanced Cognition based Framework for Design of Complex Systems

Funders :

CE - Commission Européenne [BE]

Available on ORBilu :

since 09 January 2019

Statistics

Number of views

164 (47 by Unilu)

Number of downloads

111 (9 by Unilu)

More statistics

Scopus citations^®

Scopus citations^®
without self-citations

WoS citations^™

Bibliography

D. A. Cohn, A. Ghahramani, and M. I. Jordan, "Active Learning with Statistical Models, " J. Artif. Intell. Res., vol. 4, pp. 129-145, 1996.
B. Gopalakrishnan and N. D. Sidiropoulos, "Cognitive Transmit Beamforming from Binary CSIT, " IEEE Trans. Wireless Commun., vol. 14, no. 2, pp. 895-906, Feb. 2014.
A. Tsakmalis, S. Chatzinotas, and B. Ottersten, "Centralized Power Control in Cognitive Radio Networks Using Modulation and Coding Classification Feedback, " IEEE Trans. Cognitive Commun. And Networking, vol. 2, no. 3, pp. 223-237, Sept. 2016.
U. Louche and L. Ralaivola, "From Cutting Planes Algorithms to Compression Schemes and Active Learning, " in Proc. International Joint Conf. on Neural Networks (IJCNN), Jul. 2015.
K. Chaloner and I. Verdinelli, "Bayesian Experimental Design: A Review, " Stat. Sci., vol. 10, no. 3, pp. 273-304, Aug. 1995.
I. Ben-Gal and M. Caramanis, "Sequential DOE via dynamic programming, " IIE Trans., vol. 34, no. 12, pp. 1087-1100, May 2002.
D. S. Zois, M. Levorato, and U. Mitra, "Active Classification for POMDPs: A Kalman-Like State Estimator, " IEEE Trans. Signal Process., vol. 62, no. 23, pp. 6209-6214, Dec. 2014.
A. Tsakmalis, S. Chatzinotas, and B. Ottersten, "Interference Constraint Active Learning with Uncertain Feedback for Cognitive Radio Networks, " IEEE Trans. Wireless Commun., vol. 16, no. 7, pp. 4654-4668, Jul. 2017.
M. Ono, M. Pavone, Y. Kuwata, and J. Balaram, "Chance-Constrained Dynamic Programming with Application to Risk-Aware Robotic Space Exploration, " Springer US Autonomous Robots, vol. 39, no. 4, pp. 555-571, Dec. 2015.
L. M. Lopez-Ramos, A. G. Marques, and J. Ramos, "Jointly Optimal Sensing and Resource Allocation for Multiuser Interweave Cognitive Radios, " IEEE Trans. Wireless Commun., vol. 13, no. 11, pp. 5954-5967, Nov. 2014.
D. S. Zois and U. Mitra, "Active State Tracking With Sensing Costs: Analysis of Two-States and Methods for n-States, " IEEE Trans. Signal Process., vol. 65, no. 11, pp. 2828-2843, Jun. 2017.
J. L. Williams, J. W. Fisher, and A. S. Willsky, "Approximate Dynamic Programming for Communication-Constrained Sensor Network Management, " IEEE Trans. Signal Process., vol. 55, no. 8, pp. 4300-4311, Aug. 2007.
T. Minka, A family of algorithms for approximate Bayesian inference, Ph. D. thesis, MIT, Cambridge, MA, 2001.
R. Kan and C. Robotti, "On Moments of Folded and Truncated Multivariate Normal Distributions, " Journal of Computational and Graphical Statistics, Apr. 2017.
V. N. Vapnik, The Nature of Statistical Learning Theory, Springer-Verlag, New York, 2000.
K. P. Murphy, Machine Learning: A Probabilistic Perspective, MIT Press, Cambridge, MA, USA, 2012.
A. Tsakmalis, S. Chatzinotas, and B. Ottersten, "Constrained Bayesian Active Learning of Interference Channels in Cognitive Radio Networks, " https://arxiv. org/abs/1710. 08335, 2017.
D. P. Bertsekas, Dynamic Programming and Optimal Control, Athena Scientific, Belmont, MA, 1995.