mine.bib

@techreport{dimitrakakis:gbrl,
  author = {Christos Dimitrakakis},
  title = {Monte-Carlo utility estimates for Bayesian reinforcement learning},
  institution = {EPFL},
  year = {2013},
  note = {arXiv:1303.2506},
  keywords = {reinforcement learning, Bayesian inference, stochastic gradient descent, Monte Carlo}
}
@techreport{dimitrakakis:context-models:arxiv,
  author = {Christos Dimitrakakis},
  title = {Context models on sequences of covers},
  institution = {arXiv},
  number = {arXiv:1005.2263},
  year = {2011},
  keywords = {conditional denstiy estimation, Bayesian inference}
}
@techreport{Terrorist:Bounding,
  author = {Pedro Peris-Lopez and Julio C. Hernandez-Castro and Christos Dimitrakakis and Aikaterini Mitrokotsa and Juan M. E. Tapiador},
  title = {Shedding light on {RFID} distance bounding protocols and terrorist fraud attacks},
  institution = {arXiv},
  number = {0906.4618},
  year = {2010},
  keywords = {RFID, security}
}
@article{mitrokotsa:collissions:tcj,
  author = {Aikaterini Mitrokotsa and Pedro Peris-Lopez and Christos Dimitrakakis and Serge Vaudenay },
  title = {On selecting the nonce length in distance-bounding protocols},
  year = {2013},
  journal = {The Computer Journal},
  keywords = {algorithmic analysis, martingales, security, RFID, distance bounding}
}
@inproceedings{infocom:expected-loss,
  author = {Dimitrakakis, Christos and Mitrokotsa, Aikaterini and Vaudenay, Serge},
  booktitle = {INFOCOM, 2012 Proceedings IEEE},
  title = {Expected loss bounds for authentication in constrained channels},
  year = {2012},
  month = {march},
  pages = {478--485},
  doi = {10.1109/INFCOM.2012.6195788},
  issn = {0743-166X},
  keywords = {algorithmic analysis, performance bounds, security, distance bounding}
}
@article{PLoSCB:sorn,
  author = {Pengsheng Zhang and Christos Dimitrakakis and Jochen Triesch},
  title = {Network Self-organization Explains the Statistics and Dynamics of Synaptic Connection Strengths in Cortex},
  year = {2013},
  journal = {PLoS Computational Biology},
  doi = {10.1371/journal.pcbi.1002848},
  volume = {9},
  number = {1},
  pages = {e1002848},
  keywords = {neuroscience, statistical analysis, synaptic dynamics, neural networks}
}
@inbook{lonini:imclever,
  author = {Luca Lonini and Christos Dimitrakakis and Constantin Rothkopf and Jochen Triesch},
  chapter = {Generalization and interference in human motor control},
  title = {Computational and Robotic Models of the Hierarchical Organization of Behavior},
  publisher = {Springer},
  year = {2012},
  keywords = {reinforcement learning, hierarchical architectures}
}
@inproceedings{SfN:sorn,
  author = {Pengsheng Zhang and Christos Dimitrakakis and Jochen Triesch},
  title = {Network Self-organization Explains Distribution of Synaptic Efficacies in Neocortex},
  year = {2011},
  booktiltle = {{DeveLeaNN 2011}: Workshop on Development and Learning in Artificial Neural Networks},
  keywords = {neuroscience, statistical analysis, synaptic dynamics, neural networks}
}
@techreport{expected-loss:arxiv,
  author = {Christos Dimitrakakis and Aikaterini Mitrokotsa and Serge Vaudenay},
  title = {Expected loss analysis of thresholded authentication protocols in noisy conditions},
  institution = {arXiv},
  number = {1009.0278},
  year = {2011},
  keywords = {neuroscience, statistical analysis, synaptic dynamics, neural networks}
}
@misc{TORCS,
  author = {Eric Espi\'e and Christophe Guionneau and Bernhard Wymann and Christos Dimitrakakis and R\'emi Coulom and Andrew Sumner},
  title = {{TORCS}, The Open Racing Car Simulator},
  howpublished = {\texttt{http://www.torcs.org}},
  year = {2005},
  keywords = {racing, software, robots, driving, artificial intelligence, simulation}
}
@misc{beliefbox,
  author = {Christos Dimitrakakis},
  title = {Beliefbox: A framework for statistical methods in sequential decision making},
  howpublished = {\url{http://code.google.com/p/beliefbox/}},
  year = {2007},
  keywords = {Bayesian inference, statistics, reinforcement learning}
}
@incollection{icis-book:dimitrakakis,
  author = {Christos Dimitrakakis},
  title = {Efficient methods for near-optimal sequential decision making under uncertainty},
  booktitle = {Interactive Collaborative Information Systems},
  series = {SCI},
  publisher = {Springer},
  volume = {281},
  pages = {125--153},
  year = {2010},
  editor = {Robert Babuska and Frans Groen},
  keywords = {Bayesian inference, statistics, reinforcement learning, decision theory}
}
@article{dimitrakakis:eurasip:jasmp,
  author = {Christos Dimitrakakis and Samy Bengio},
  title = {Phoneme and Sentence-Level Ensembles for Speech Recognition},
  journal = {{EURASIP} Journal on Audio, Speech and Music Processing},
  year = {2011},
  keywords = {speech recognition, ensembles}
}
@article{mafia-fraud,
  author = {Aikaterini Mitrokotsa and Christos Dimitrakakis and Pedro Peris-Lopez and Juan C. Hernandez-Castro},
  title = {Reid et al.'s Distance Bounding Protocol and Mafia Fraud Attacks over Noisy Channels},
  journal = {IEEE Communication Letters},
  year = {2010},
  volume = {14},
  number = {2},
  pages = {121--123},
  keywords = {RFID, security, distance bounding, algorthmic analysis}
}
@article{dimitrakakis+lagoudakis:mlj2008,
  author = {Christos Dimitrakakis and Michail G. Lagoudakis},
  title = {Rollout Sampling Approximate Policy Iteration},
  journal = {Machine Learning},
  volume = {72},
  number = {3},
  pages = {157--171},
  month = {September},
  year = {2008},
  doi = {10.1007/s10994-008-5069-3},
  note = {Presented at ECML'08},
  keywords = {reinforcement learnig, policy iteration, rollout methods}
}
@article{dimitrakakis08ogai,
  author = {Christos Dimitrakakis},
  title = {Exploration in {POMDP}s},
  journal = {\"{O}sterreichische Gesellschaft f\"{u}r Artificial Intelligence Journal},
  year = {2008},
  volume = {1},
  pages = {24--31},
  keywords = {reinforcement learning, Bayesian, partial observability, exploration-expoloitation}
}
@article{neurocomputing:Dimitrakakis+Bengio:2005,
  author = {Christos Dimitrakakis and Samy Bengio},
  title = {Online Policy Adaptation for Ensemble Classifiers},
  journal = {Neurocomputing},
  year = 2005,
  volume = 64,
  pages = {211--221},
  keywords = {reinforcement learning, ensemble methods, classification}
}
@techreport{dimitrakakis:srp:arxiv,
  author = {Christos Dimitrakakis},
  title = {Sparse Reward Processes},
  institution = {arXiv},
  number = {abs/1201.2555},
  year = {2012},
  note = {Presented at Dagstuhl Seminar on Machine Learning and Security},
  keywords = {Reinforcement learning, multi-task learning, curiosity, security}
}
@inproceedings{dimitrakakis:mmbi:ewrl:2011,
  author = {Christos Dimitrakakis},
  title = {Robust Bayesian reinforcement learning through tight lower bounds},
  booktitle = {European Workshop on Reinforcement Learning (EWRL 2011)},
  year = {2011},
  series = {LNCS},
  number = {7188},
  pages = {177--188},
  keywords = {Reinforcement learning, Bayesian inference, Thompson sampling, value function bounds, complexity}
}
@inproceedings{dimitrakakis:bmirl:ewrl:2011,
  author = {Christos Dimitrakakis and Constantin A. Rothkopf},
  title = {Bayesian Multitask Inverse Reinforcement Learning},
  booktitle = {European Workshop on Reinforcement Learning (EWRL 2011)},
  year = {2011},
  series = {LNCS},
  number = {7188},
  pages = {273--284},
  keywords = {Inverse reinforcement learning, apprenticeship learning, Bayesian inference, multi-task learning, complexity, behavioural modelling}
}
@inproceedings{rothkopf:peirl:ecml:2011,
  author = {Constantin A. Rothkopf and Christos Dimitrakakis},
  title = {Preference Elicitation and Inverse Reinforcement Learning},
  booktitle = {ECML/PKDD (3)},
  year = {2011},
  series = {LNCS},
  volume = {6913},
  pages = {34-48},
  keywords = {Inverse reinforcement learning, apprenticeship learning, Bayesian inference,  behavioural modelling}
}
@inproceedings{brammert:duct:aaai:2012,
  author = {Brammert Ottens and Christos Dimitrakakis and Boi Faltings},
  title = {{DUCT}: An Upper Confidence Bound Approach to Distributed Constraint Optimization Problems},
  year = {2012},
  booktitle = {{AAAI} 2012},
  keywords = {distributed constrained optimisation, confidence bounds, complexity}
}
@inproceedings{dimitrakakis:icml:2010,
  author = {Christos Dimitrakakis},
  title = {Context model inference for large or partially observable {MDPs}},
  booktitle = {{ICML} workshop on reinforcement learning and search in very large spaces},
  year = {2010},
  keywords = {Reinforcement learning, Bayesian inference, approximate dynamic programming}
}
@inproceedings{dimitrakakis:aistats:2010,
  author = {Christos Dimitrakakis},
  title = {{Bayesian} Variable Order {Markov} Models},
  booktitle = {Proceedings of the 13th International Conference on Artificial Intelligence and Statistics ({AISTATS})},
  pages = {161--168},
  year = {2010},
  editor = {Yee Whye Teh and Mike Titterington},
  volume = {9},
  series = {{JMLR : W\&CP}},
  address = {Chia Laguna Resort, Sardinia, Italy},
  keywords = {Bayesian inference, prediction}
}
@inproceedings{dimitrakakis-mitrokotsa:icmla2009,
  author = {Christos Dimitrakakis and Aikaterini Mitrokotsa},
  title = {Statistical Decision Making for Authentication and Intrusion Detection},
  booktitle = {Machine Learning and Applications, Fourth International Conference on ({ICMLA'09})},
  isbn = {978-0-7695-3926-3},
  year = {2009},
  pages = {409--414},
  doi = {http://doi.ieeecomputersociety.org/10.1109/ICMLA.2009.46},
  publisher = {IEEE Computer Society},
  address = {Miami, FL, USA},
  keywords = {Bayesian inference, minimax priors, security}
}
@inproceedings{dimitrakakis:icaart2010,
  author = {Christos Dimitrakakis},
  title = {Complexity of stochastic branch and bound methods for belief tree search in {Bayesian} reinforcement learning},
  booktitle = {2nd international conference on agents and artificial intelligence ({ICAART 2010})},
  address = {Valencia, Spain},
  organization = {ISNTICC},
  publisher = {Springer},
  pages = {259--264},
  year = {2010},
  notes = {arXiv:0912.5029},
  keywords = {Bayesian inference, reinforcement learning, exploration-exploitation, planning, stochastic branch and bound, Thompson sampling, value function bounds, complexity}
}
@inproceedings{dimitrakakis:cimca08,
  author = {Christos Dimitrakakis},
  title = {Tree Exploration for {Bayesian} {RL} Exploration},
  booktitle = {Computational Intelligence for Modelling, Control and Automation, International Conference on},
  year = {2008},
  isbn = {978-0-7695-3514-2},
  pages = {1029-1034},
  doi = {http://doi.ieeecomputersociety.org/10.1109/CIMCA.2008.32},
  publisher = {IEEE Computer Society},
  address = {Wien, Austria},
  keywords = {Bayesian inference, reinforcement learning, exploration-exploitation, planning, Thompson sampling, value function bounds}
}
@inproceedings{dimitrakakis+lagoudakis:ewrl2008,
  author = {Christos Dimitrakakis and
               Michail G. Lagoudakis},
  title = {Algorithms and Bounds for Rollout Sampling Approximate Policy
               Iteration},
  booktitle = {EWRL},
  year = {2008},
  pages = {27-40},
  ee = {http://dx.doi.org/10.1007/978-3-540-89722-4_3},
  bibsource = {DBLP, http://dblp.uni-trier.de},
  keywords = {Reinforcement learning, rollout algorithms, resource allocation, approximate policy iteration}
}
@inproceedings{dimitrakakis2008foiks,
  title = {Cost-minimising strategies for data labelling: optimal stopping and active learning},
  author = {Christos Dimitrakakis and Christian Savu-Krohn},
  booktitle = {Proceedings of the 5th international symposium on Foundations of Information and Knowledge Systems ({FoIKS} 2008)},
  series = {Lecture Notes in Computer Science},
  volume = {4932},
  pages = {96-111},
  year = {2008},
  publisher = {Springer},
  address = {Pisa, Italy},
  month = feb,
  keywords = {active learning, Bayesian inference, optimal stopping}
}
@inproceedings{ec2nd:mitrokotsa:2007,
  address = {Heraklion, Greece},
  author = {A. Mitrokotsa and C. Dimitrakakis and C. Douligeris},
  booktitle = {Proceedings of the 3rd European Conference on Computer Network Defense (EC2ND'07) },
  month = {4-5 October},
  pages = {35--46},
  publisher = {Springer},
  series = {LNEE (Lecture Notes in Electrical Engineering)},
  volume = 30,
  title = {Intrusion Detection Using Cost-Sensitive Classification},
  year = {2007},
  keywords = {classification, security}
}
@article{mitrokotsa:manet:adhoc,
  author = {Aikaterini Mitrokotsa and Christos Dimitrakakis},
  title = {Intrusion Detection in MANET using classification Algorithms: The Effects of Cost and Model Selection},
  journal = {Ad-Hoc Networks},
  year = {2012},
  keywords = {classification, security}
}
@inproceedings{dimitrakakis:icann:2006,
  booktitle = {Int. Conf. on Artificial Neural Networks ({ICANN})},
  author = {Christos Dimitrakakis},
  title = {Nearly optimal exploration-exploitation decision thresholds},
  year = {2006},
  abstract = {While in general trading off exploration and exploitation in
reinforcement learning is hard, under some formulations relatively simple solutions exist. Optimal decision thresholds for the multi-armed bandit problem, one for the infinite horizon discounted reward case and one for the finite horizon undiscounted reward case are derived, which make the link between the reward horizon, uncertainty and the need for exploration explicit. From this result follow two practical approximate algorithms related to Thompson sampling, which are illustrated experimentally.},
  ipdmembership = {learning},
  ipdpostscript = {ftp://ftp.idiap.ch/pub/papers/2006/dimitrakakis-icann-2006.ps.gz},
  ipdpdf = {ftp://ftp.idiap.ch/pub/papers/2006/dimitrakakis-icann-2006.pdf},
  ipdxref = {techreport:dimitrakakis-idiap-rr-06-12.bib},
  keywords = {reinforcement learning, exploration-exploitation, Thompson sampling, confidence bounds, ensemble methods}
}
@inproceedings{pascal:Dimitrakakis+Bengio:2005,
  author = {Christos Dimitrakakis and Samy Bengio},
  title = {Gradient-based estimates of return distributions},
  booktitle = {PASCAL workshop on principled methods of trading exploration and exploitation},
  year = {2005},
  organization = {PASCAL Network},
  keywords = {reinforcement learning, gradient descent, exploration-exploitation}
}
@inproceedings{icassp:Dimitrakakis+Bengio:2004,
  author = {Christos Dimitrakakis and Samy Bengio},
  title = {Boosting {HMM}s with an application to speech recognition},
  year = {2004},
  booktitle = {Proceedings of the {IEEE} International Conference on Acoustic, Speech, and Signal Processing ({ICASSP})},
  volume = {5},
  pages = {621--624},
  abstract = {Boosting is a general method for training an ensemble of
classifiers with a view to improving performance relative to that of a
single classifier. While the original AdaBoost algorithm has been
defined for classification tasks, the current work examines its
applicability to sequence learning problems. In particular, different
methods for training HMMs on sequences and for combining their output
are investigated in the context of automatic speech recognition.},
  ipdmembership = {learning},
  ipdpostscript = {http://www.idiap.ch/~dimitrak/downloads/dimitrak_bengio03b.ps.gz},
  ipdpdf = {http://www.idiap.ch/~dimitrak/downloads/dimitrak_bengio03b.pdf},
  idxref = {techreport:dimitrak-rr-03-41.bib},
  keywords = {speech recognition, ensemble methods, hidden Markov models}
}
@inproceedings{icassp:Dimitrakakis+Bengio:2005,
  author = {Christos Dimitrakakis and Samy Bengio},
  title = {Boosting word error rates},
  year = 2005,
  booktitle = {Proceedings of the {IEEE} International Conference on Acoustic, Speech, and Signal Processing ({ICASSP})},
  volume = {5},
  pages = {501--504},
  abstract = {We apply boosting techniques to the problem of word error
  rate minimisation in speech recognition. This is achieved through a
  new definition of sample error for boosting and a training procedure
  for hidden Markov models. For this purpose we define a sample error
  for sentence examples related to the word error rate. Furthermore,
  for each sentence example we define a probability distribution in
  time that represents our belief that an error has been made at that
  particular frame. This is used to weigh the frames of each sentence
  in the boosting framework. We present preliminary results on the
  well-known Numbers 95 database that indicate the importance of this
  temporal probability distribution.},
  ipdmembership = {learning},
  ipdpostscript = {http://www.idiap.ch/~dimitrak/papers/dimitrak_bengio_04b.ps.gz},
  ipdpdf = {http://www.idiap.ch/~dimitrak/papers/dimitrak_bengio_04b.pdf},
  ipdxref = {techreport:dimitrak-rr-04-49.bib},
  keywords = {speech recognition, ensemble methods, hidden Markov models}
}
@inproceedings{dimitrakakis2004opa,
  title = {{Online Policy Adaptation for Ensemble Classifiers}},
  author = {Christos Dimitrakakis and Samy Bengio},
  booktitle = {12th European Symposium on Artificial Neural Networks, {ESANN}},
  volume = {4},
  year = {2004},
  keywords = {reinforcement learning, ensemble methods, classification}
}
@techreport{dimitrakakis:tr-fias-10-01,
  author = {Christos Dimitrakakis},
  title = {Variable order {Markov} decision processes: Exact {Bayesian} inference with an application to {POMDPs}},
  year = {2010},
  institution = {FIAS},
  month = {May},
  note = {http://fias.uni-frankfurt.de/$\sim$dimitrakakis/papers/tr-fias-10-05.pdf},
  keywords = {POMDPs, variable order Markov model, Bayesian inference, approximate dynamic programming}
}
@techreport{dimitrakakis:tr-uva-09-04,
  author = {Christos Dimitrakakis},
  title = {Bayesian Variable Order {Markov} Models: Towards {Bayesian} Predictive State Representations},
  year = {2009},
  number = {IAS-UVA-09-04},
  institution = {University of Amsterdam},
  month = {June},
  keywords = {POMDPs, variable order Markov model, Bayesian inference, approximate dynamic programming, predictive state representations}
}
@techreport{dimitrakakis:tr-uva-09-01,
  author = {Christos Dimitrakakis},
  title = {Complexity of stochastic branch and bound for belief tree search in {Bayesian} reinforcement learning},
  institution = {University of Amsterdam},
  year = {2009},
  number = {IAS-UVA-09-01},
  month = {April}
}
@techreport{dimitrakakis:rr05-29,
  author = {Christos Dimitrakakis and Samy Bengio},
  title = {Gradient estimates of return},
  institution = {IDIAP},
  year = {2005},
  type = {IDIAP-RR},
  number = {05-29},
  abstract = {The exploration-exploitation trade-off that arises when one considers simple point estimates of expected returns no longer appears when full distributions are considered. This work develops a simple gradient-based approach for mainting such distributions and investigates methods for using them to direct exploration.},
  ipdmembership = {learning},
  ipdpostscript = {ftp://ftp.idiap.ch/pub/reports/2005/dimitrakakis-idiap-rr-05-29.ps.gz},
  ipdpdf = {ftp://ftp.idiap.ch/pub/reports/2005/dimitrakakis-idiap-rr-05-29.pdf},
  ipdxref = {article:dimitrakakis-pascal-2005.bib}
}
@techreport{dimitrak+bengio:04-72,
  author = {Christos Dimitrakakis and Samy Bengio},
  title = {Estimates of Parameter Distributions for Optimal Action Selection},
  institution = {IDIAP},
  year = 2004,
  number = {04-72},
  abstract = {We present a general method for maintaining estimates of
the distribution of parameters in arbitrary models. This is then
applied to the estimation of probability distribution over actions in
value-based reinforcement learning. While this approach is similar to
other techniques that maintain a confidence measure for action-values,
it nevertheless offers a new insight into current techniques and
reveals potential avenues of further research.},
  ipdmembership = {learning},
  ipdpostscript = {ftp://ftp.idiap.ch/pub/reports/2004/rr-04-72.ps.gz},
  ipdpdf = {ftp://ftp.idiap.ch/pub/reports/2004/rr-04-72.pdf}
}
@techreport{dimitrakakis:rr06-13,
  author = {Christos Dimitrakakis},
  title = {Online statistical estimation for vehicle control},
  institution = {IDIAP},
  year = {2006},
  type = {IDIAP-RR},
  number = {13},
  abstract = {This tutorial examines simple physical models of vehicle dynamics and overviews methods for parameter estimation and control. Firstly, techniques for the estimation of parameters that deal with constraints are detailed. Secondly, methods for controlling the system are explained.},
  ipdmembership = {learning},
  ipdpostscript = {ftp://ftp.idiap.ch/pub/reports/2006/dimitrakakis-idiap-rr-06-13.ps.gz},
  ipdpdf = {ftp://ftp.idiap.ch/pub/reports/2006/dimitrakakis-idiap-rr-06-13.pdf}
}
@misc{unpublished:Dimitrakakis,
  author = {Christos Dimitrakakis},
  title = {Reinforcement Learning with Continuous Action Values},
  note = {http://christos.dimitrakakis.googlepages.com/RLContAction.ps.gz},
  year = {1999}
}
@phdthesis{christos-dimitrakakis:phd-thesis:2006,
  author = {Christos Dimitrakakis},
  title = {Ensembles for Sequence Learning},
  school = {{\'E}cole Polytechnique F{\'e}d{\'e}rale de Lausanne},
  year = {2006},
  ipdmembership = {Learning},
  ipdpostscript = {ftp://ftp.idiap.ch/pub/reports/2006/christos-dimitrakakis-phd-thesis.ps.gz},
  ipdpdf = {ftp://ftp.idiap.ch/pub/reports/2006/christos-dimitrakakis-phd-thesis.pdf},
  keywords = {Ensembles, boosting, bagging, mixture of experts, speech
recognition, reinforcement learning, exploration-exploitation, uncertainty,
sequence learning, sequential decision making, Thompson sampling},
  abstract = {This thesis explores the application of ensemble methods to sequential
learning tasks. The focus is on the development and the critical
examination of new methods or novel applications of existing methods,
with emphasis on supervised and reinforcement learning problems.

In both types of problems, even after having observed a certain amount
of data, we are often faced with uncertainty as to which hypothesis is
correct among all the possible ones.  However, in many methods for
both supervised and for reinforcement learning problems this
uncertainty is ignored, in the sense that there is a single solution
selected out of the whole of the hypothesis space.  Apart from the
classical solution of analytical Bayesian formulations, ensemble
methods offer an alternative approach to representing this
uncertainty.  This is done simply through maintaining a set of
alternative hypotheses.

The sequential supervised problem considered is that of automatic
speech recognition using hidden Markov models.  The application of
ensemble methods to the problem represents a challenge in itself,
since most such methods can not be readily adapted to sequential
learning tasks.  This thesis proposes a number of different approaches
for applying ensemble methods to speech
recognition and develops methods for effective training of phonetic
mixtures with or without access to phonetic alignment data.
Furthermore, the notion of expected loss is introduced for integrating
probabilistic models with the boosting approach.  In
some cases substantial improvements over the baseline system are
obtained.

In reinforcement learning problems the goal is to act in such a way as
to maximise future reward in a given environment.  In such problems
uncertainty becomes important since neither the environment nor the
distribution of rewards that result from each action are known.  This
thesis presents novel algorithms for acting nearly optimally under
uncertainty based on theoretical considerations.  Some ensemble-based
representations of uncertainty (including a fully Bayesian model) are
developed and tested on a few simple tasks resulting in performance
comparable with the state of the art.  The thesis also draws some
parallels between a proposed representation of uncertainty based on
gradient-estimates and on ``prioritised sweeping'' and between the
application of reinforcement learning to controlling an ensemble of
classifiers and classical supervised ensemble learning methods.
}
}
@inproceedings{recsys:vmm,
  author = {Florent Garcin and Christos Dimitrakakis and Boi Faltings},
  title = {Variable Order Markov Model Recommender Systems for Personalized News Recommendation},
  year = {2012},
  keywords = {variable order Markov model, recommender system, Bayesian inference}
}
@inproceedings{icml:irl,
  author = {Aristide Tossou and Christos Dimitrakakis},
  title = {Probabilistic inverse reinforcement learning in unkown environments},
  note = {submitted},
  year = {2013},
  keywords = {probabilistic inference, inverse reinforcement learning, apprenticeship learning}
}
@inproceedings{ijcai:lbrl,
  author = {Nikolaos Tziortziotis and Christos Dimitrakakis and Konstantinos Blekas},
  title = {Linear Bayesian Reinforcement Learning},
  booktitle = {Proceedings of the 23rd international joint conference on artififical intelligence ({IJCAI} 2013)},
  year = {2013},
  keywords = {Bayesian inference, approximate dynamic programming, Thompson sampling}
}
@inproceedings{uai:ctbrl,
  author = {Nikolaos Tziortziotis and Christos Dimitrakakis and Konstantinos Blekas},
  title = {Cover Tree Bayesian Reinforcement Learning},
  note = {submitted},
  year = {2013},
  keywords = {cover trees, non-parametric Bayesian models, Bayesian inference, approximate dynamic programming, Thompson sampling}
}
@inproceedings{icml:abcrl,
  author = {Christos Dimitrakakis and Nikolaos Tziortziotis},
  title = {{ABC} Reinforcement Learning},
  note = {submitted},
  year = {2013},
  keywords = {Bayesian analysis, approximate Bayesian computation, approximate dynamic programming, Thompson sampling}
}

This file was generated by bibtex2html 1.95.