Aktivní učení v regresních modelech

• Název práce v češtině: Aktivní učení v regresních modelech
• Název v anglickém jazyce: Actve lerning in regression models
• Klíčová slova: regresní modely, umělé neuronové sítě, náhradní modely, časové řady, aktivní učení, semi-supervizované učení
• Klíčová slova anglicky: regression models, artificial neural networks, surrogate models, time series, active learning semi-supervized learning
• Akademický rok vypsání: 2016/2017
• Typ práce: disertační práce
• Jazyk práce: čeština
• Ústav: Katedra teoretické informatiky a matematické logiky (32-KTIML)
• Vedoucí / školitel: prof. RNDr. Ing. Martin Holeňa, CSc.
• Řešitel: skrytý - zadáno a potvrzeno stud. odd.
• Datum přihlášení: 20.09.2017
• Datum zadání: 20.09.2017
• Datum potvrzení stud. oddělením: 03.10.2017

Zásady pro vypracování

The prospective PhD student gets familiar with the usual way of applying active learning to classification, including the possibility to combine active and semi-supervised learning. Then he concentrates on the less investigated area of applying active learning to regression, including its application to artificial neural networks, surrogate models in black-box optimization, and to time series.

Seznam odborné literatury

* S. Abdullah, J.C. Allwright. An active learning approach for radial basis function neural networks. Jurnal Teknologi, 45 (2006): 77–96.
* T. Alpcan. A framework for optimization under limited information. J Glob Optim 55 (2013):6 81–706.
* B. Anderson, A. Moore. Active Learning for Hidden Markov Models: Objective Functions and Algorithms, ICML 2005, 9-16.
* P. Balaprakash, R.B. Gramacy, S.M. Wild. Active-Learning-Based Surrogate Models for Empirical Performance Tuning, IEEE Cluster Computing 2013.
* I. Couckuyt, D. Gorissen, K. Crombecq, D. Deschrijver, T. Dhaene. The SUMO Toolbox: a Tool for Automatic
Regression Modeling and Active Learning. AFRICON 2013.
* D. Gorrissen. Grid-enabled Adaptive Surrogate Modeling for Computer Aided Engineering. Dissertation, 2010.
* A. Krause, C. Guestrin.Nonmyopic Active Learning of Gaussian Processes: An Exploration–Exploitation Approach. ICML 2007, 449-456.
* C.H. Lin, M. Mausam, D.S. Weld. Re-Active Learning: Active Learning with Relabeling, AAAI 2016, 1845-1852.
* Z. Lu, X. Wu, J.C. Bongard. Active Learning through Adaptive Heterogeneous Ensembling, IEEE Transactions on Knowledge and Data Engineering 27 (2015): 368-381.
* T. Schaul. Studies in Continuous Black-box Optimization. Dissertation, 2011.
* B. Settles. An Analysis of Active Learning Strategies for Sequence Labeling Tasks, EMNLP 2008, 1070-1078.
* B. Settles. Active Learning Literature Survey. TR Universitz of Wisconsin-Madison, 2010.
* M. Sugiyama, N. Rubens. A batch ensemble approach to active learning with model selection. Neural Networks, 21 (2008): 1278–1286.
* K. Tomanek, U. Hahn. Semi-Supervised Active Learning for Sequence Labeling, IJCNLP 2009, 1039-1047.
* K. Yu, J. Bi, V. Tresp. Active Learning via Transductive Experimental Design. ICML 2006, 1081-1088.
* L. Zhang, C. Chen, J. Bu, D. Ca, X. He, T. Huang. Active Learning based on Locally Linear Reconstruction. IEEE Transactions on Pattern Analysis and Machine Intelligence, 33 (2011): 2026-2038
* S. Zhou, Q. Chen, X. Wang. Active Semi-Supervised Learning Method with Hybrid Deep Belief Networks, Plos One, 9 (2014): e107122