Positions

Research Areas research areas

Overview

  • Personal page: http://cs.unh.edu/~mpetrik
  • Publications

    Academic Article

    Year Title
    2015 Tight Approximations of Dynamic Risk MeasuresMathematics of Operations Research.  40:655-682. 2015
    2014 Social media and customer behavior analytics for personalized customer engagementsIBM Journal of Research and Development.  58:7:1-7:12. 2014
    2014 Efficient and accurate methods for updating generalized linear models with multiple feature additionsJournal of Machine Learning Research.  15:2607-2627. 2014
    2011 Robust approximate bilinear programming for value function approximationJournal of Machine Learning Research.  12:3027-3063. 2011
    2009 Hybrid least-squares algorithms for approximate policy evaluationMachine Learning.  76:243-256. 2009
    2009 A bilinear programming approach for multiagent planningJournal of Artificial Intelligence Research.  35:235-274. 2009
    2007 Learning parallel portfolios of algorithmsAnnals of Mathematics and Artificial Intelligence.  48:85-106. 2007
    2005 Learning parallel portfolios of algorithms 2005

    Conference Proceeding

    Year Title
    2018 Fast Bellman Updates for Robust MDPs 2018
    2018 Low-rank Feature Selection for Reinforcement Learning 2018
    2017 A Practical Method for Solving Contextual Bandit Problems Using Decision TreesUncertainty in Artificial Intelligence (UAI). 2017
    2017 Robust Partially-Compressed Least-Squares Robust Partially-Compressed Least-SquaresAAAI Conference on Artificial Intelligence. 2017
    2017 Value Directed Exploration in Multi-Armed Bandits with Structured PriorsUncertainty in Artificial Intelligence (UAI). 1-16. 2017
    2017 Value Directed Exploration in Multi-Armed Bandits with Structured Priors.UAI. 2017
    2016 Building an Interpretable Recommender via Loss-Preserving TransformationCML Workshop on Human Interpretability in Machine Learning. 2016
    2016 Interpretable Policies for Dynamic Product RecommendationsUncertainty in Artificial Intelligence (UAI). 2016
    2016 Safe Policy Improvement by Minimizing Robust Baseline RegretICML Workshop on Reliable Machine Learning in the Wild. 1-25. 2016
    2016 Safe Policy Improvement by Minimizing Robust Baseline RegretAdvances in Neural Information Processing Systems. 2016
    2015 Optimal Threshold Control for Energy Arbitrage with Degradable Battery StorageUncertainty in Artificial Intelligence (UAI). 692-701. 2015
    2014 RAAM : The benefits of robustness in approximating aggregated MDPs in reinforcement learningNeural Information Processing Systems (NIPS). 2014
    2013 Agile logistics simulation and optimization for managing disaster responses.Winter Simulation Conference. 3340-3351. 2013
    2013 Solution methods for constrained Markov decision process with continuous probability modulationUncertainty in Artificial Intelligence (UAI). 2013
    2012 An approximate solution method for large risk-averse Markov decision processesUncertainty in Artificial Intelligence (UAI). 2012
    2012 Approximate dynamic programming by minimizing distributionally robust boundsInternational Conference of Machine Learning. 2012
    2012 Optimizing the end-to-end value chain through demand shaping and advanced customer analyticsSymposium on Process Systems Engineering. 15-19. 2012
    2011 Linear dynamic programs for resource managementConference on Artificial Intelligence (AAAI). 2011
    2010 Feature selection using regularization in approximate linear programs for Markov decision processesInternational Conference on Machine Learning. 2010
    2009 Constraint relaxation in approximate linear programsInternational Conference on Machine Learning. 2009
    2009 Robust value function approximation using bilinear programmingAdvances in Neural Information Processing Systems (NIPS). 2009
    2008 A successive approximation algorithm for coordination problemsISAIM. 2008
    2008 Biasing approximate dynamic programming with a lower discount factorAdvances in Neural Information Processing Systems (NIPS). 2008
    2008 Interaction structure and dimensionality in decentralized problem solvingConference on Artificial Intelligence (AAAI). 1440-1441. 2008
    2008 Learning heuristic functions through approximate linear programmingInternational Conference on Automated Planning and Scheduling (ICAPS). 2008
    2007 An analysis of Laplacian methods for value function approximation in MDPsInternational Joint Conference on Artificial Intelligence. 2574-2579. 2007
    2007 Anytime coordination using separable bilinear programsConference on Artificial Intelligence. 750-755. 2007
    2007 Average-reward decentralized Markov decision processesInternational Joint Conference on Artificial Intelligence. 1997-2002. 2007
    Computing Robust Strategies for Managing Invasive Plants

    Teaching Activities

  • Doctoral Research Taught course 2018
  • Topics/Machine Learning Taught course 2018
  • Independent Study Taught course 2018
  • Doctoral Research Taught course 2018
  • Independent Study Taught course 2018
  • Rsrch Exp/Computer Science Taught course 2018
  • Top/Machine Learning Taught course 2018
  • Independent Study/Honors Taught course 2017
  • Topics/Machine Learning Taught course 2017
  • Independent Study Taught course 2017
  • Top/Intro to Machine Learning Taught course 2017
  • Topics/Machine Learning Taught course 2016
  • Full Name

  • Marek Petrik