Machine Learning and Knowledge Discovery in Databases

1. Front Matter

   PDF

2. Reinforcement Learning

   1. A Cascaded Supervised Learning Approach to Inverse Reinforcement Learning

      • Edouard Klein, Bilal Piot, Matthieu Geist, Olivier Pietquin

      Pages 1-16

   2. Learning from Demonstrations: Is It Worth Estimating a Reward Function?

      • Bilal Piot, Matthieu Geist, Olivier Pietquin

      Pages 17-32

   3. Recognition of Agents Based on Observation of Their Sequential Behavior

      • Qifeng Qiao, Peter A. Beling

      Pages 33-48

   4. Learning Throttle Valve Control Using Policy Search

      • Bastian Bischoff, Duy Nguyen-Tuong, Torsten Koller, Heiner Markert, Alois Knoll

      Pages 49-64

   5. Model-Selection for Non-parametric Function Approximation in Continuous Control Problems: A Case Study in a Smart Energy System

      • Daniel Urieli, Peter Stone

      Pages 65-80

   6. Learning Graph-Based Representations for Continuous Reinforcement Learning Domains

      • Jan Hendrik Metzen

      Pages 81-96

   7. Regret Bounds for Reinforcement Learning with Policy Advice

      • Mohammad Gheshlaghi Azar, Alessandro Lazaric, Emma Brunskill

      Pages 97-112

   8. Exploiting Multi-step Sample Trajectories for Approximate Value Iteration

      • Robert Wright, Steven Loscalzo, Philip Dexter, Lei Yu

      Pages 113-128

3. Markov Decision Processes

   1. Expectation Maximization for Average Reward Decentralized POMDPs

      • Joni Pajarinen, Jaakko Peltonen

      Pages 129-144

   2. Properly Acting under Partial Observability with Action Feasibility Constraints

      • Caroline P. Carvalho Chanel, Florent Teichteil-Königsbuch

      Pages 145-161

   3. Iterative Model Refinement of Recommender MDPs Based on Expert Feedback

      • Omar Zia Khan, Pascal Poupart, John Mark Agosta

      Pages 162-177

   4. Solving Relational MDPs with Exogenous Events and Additive Rewards

      • Saket Joshi, Roni Khardon, Prasad Tadepalli, Aswin Raghavan, Alan Fern

      Pages 178-193

   5. Continuous Upper Confidence Trees with Polynomial Exploration – Consistency

      • David Auger, Adrien Couëtoux, Olivier Teytaud

      Pages 194-209

4. Active Learning and Optimization

   1. A Lipschitz Exploration-Exploitation Scheme for Bayesian Optimization

      • Ali Jalali, Javad Azimi, Xiaoli Fern, Ruofei Zhang

      Pages 210-224

   2. Parallel Gaussian Process Optimization with Upper Confidence Bound and Pure Exploration

      • Emile Contal, David Buffoni, Alexandre Robicquet, Nicolas Vayatis

      Pages 225-240

   3. Greedy Confidence Pursuit: A Pragmatic Approach to Multi-bandit Optimization

      • Philip Bachman, Doina Precup

      Pages 241-256

   4. A Time and Space Efficient Algorithm for Contextual Linear Bandits

      • José Bento, Stratis Ioannidis, S. Muthukrishnan, Jinyun Yan

      Pages 257-272

   5. Knowledge Transfer for Multi-labeler Active Learning

      • Meng Fang, Jie Yin, Xingquan Zhu

      Pages 273-288

5. Learning from Sequences

   1. Spectral Learning of Sequence Taggers over Continuous Sequences

      • Adrià Recasens, Ariadna Quattoni

      Pages 289-304

This three-volume set LNAI 8188, 8189 and 8190 constitutes the refereed proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases, ECML PKDD 2013, held in Prague, Czech Republic, in September 2013. The 111 revised research papers presented together with 5 invited talks were carefully reviewed and selected from 447 submissions. The papers are organized in topical sections on reinforcement learning; Markov decision processes; active learning and optimization; learning from sequences; time series and spatio-temporal data; data streams; graphs and networks; social network analysis; natural language processing and information extraction; ranking and recommender systems; matrix and tensor analysis; structured output prediction, multi-label and multi-task learning; transfer learning; bayesian learning; graphical models; nearest-neighbor methods; ensembles; statistical learning; semi-supervised learning; unsupervised learning; subgroup discovery, outlier detection and anomaly detection; privacy and security; evaluation; applications; and medical applications.

• bayesian network
• data mining
• graph-based methods
• parallel optimization
• social responsibility

• Department of Computer Science, Katholieke Universiteit Leuven, Leuven, Belgium

  Hendrik Blockeel

• Fraunhofer IAIS, Department of Knowledge Discovery, University of Bonn, Sankt Augustin, Germany

  Kristian Kersting

• LIACS, Universiteit Leiden, Leiden, The Netherlands

  Siegfried Nijssen

• Department of Computer Science and Engineering, Czech Technical University, Prague 6, Czech Republic

  Filip Železný

Policies and ethics