Michail G. Lagoudakis [www.intelligence.tuc.gr/~lagoudakis] is currently a full Professor with the School of Electrical and Computer Engineering (ECE) at the Technical University of Crete (TUC), Greece since September 2005. Prior to this appointment, he held a postdoctoral fellow position at the Georgia Institute of Technology, USA, from 2003 to 2005, and a visiting researcher position at the Shannon Laboratory of AT&T Labs during the summer of 2000. He received his Doctoral degree in Computer Science with distinction from Duke University, USA in 2003, a Master’s degree in Computer Science from the University of Louisiana, Lafayette, USA in 1998, and a Diploma in Computer Engineering and Informatics from the University of Patras, Greece in 1995. His research interests are mainly in Machine Learning and Robotics with extensive research experience and expertise in the area of Decision Making under Uncertainty and Reinforcement Learning. His published research work includes 5 book chapters, 13 journal articles and 47 conference articles and has been cited more than 5300 times by other researchers (Google Scholar h-index: 29). He has served as a member on more than 40 program committees of various international conferences and as a reviewer on more than 25 scientific journals and 25 international conferences. He teaches both undergraduate (Theory of Computation, Structured Programming, Artificial Intelligence, Autonomous Agents) and graduate (Machine Learning, Probabilistic Robotics) courses at TUC and has supervised 1 doctoral, 6 master’s and 81 diploma theses. Soon after his arrival at TUC, he founded Team “Kouretes” [www.kouretes.gr], which conducts research in autonomous robotics, but also organizes and supports educational robotics activities for students of all levels of education. With the team “Kouretes” he has been continuously participating since 2013 in the annual open STEAM events of the Technical University of Crete: Science and Technology Day every autumn and Open Day every spring. He served in the Administration of the Technical University of Crete for more than five years, initially as Vice Rector for Administrative, Academic, and Student Affairs from December 2017 to August 2022 and subsequently as Rector from September 2022 to early January 2023.
Selected Publications [Scholar, ORCID, Scopus]
Fotios Lygerakis, Vassilios Diakoloulas, Michail Lagoudakis, Kotti Margarita, “Robust belief state space representation for statistical dialogue managers using deep autoencoders”, Proceedings of the IEEE Automatic Speech Recognition and Understanding Workshop (ASRU 2019), December 2019, Singapore.
Margarita Kotti, Vassilios Diakoloukas, Alexandros Papangelis, Michail G. Lagoudakis, and Yannis Stylianou, “A Case Study on the Importance of Belief State Representation for Dialogue Policy Management,” Proceedings of Interspeech 2018, Hyberabad, India, September 2018, pp. 986–990.
Michail G. Lagoudakis, “Value Function Approximation,” in Claude Sammut and Geoffrey I. Webb (Eds.), Encyclopedia of Machine Learning and Data Mining, Springer, 2014, pp. 1–15.
Michail G. Lagoudakis, “Least-Squares Reinforcement Learning Methods,” in Claude Sammut and Geoffrey I. Webb (Eds.), Encyclopedia of Machine Learning and Data Mining, Springer, 2014, pp. 1–9.
Stylianos Piperakis, Emmanouil Orfanoudakis, Michail G. Lagoudakis, “Predictive Control for Dynamic Locomotion of Real Humanoid Robots”, Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Chicago, USA, September 2014.
Ioannis Rexakis and Michail G. Lagoudakis, “Directed Policy Search for Decision Making using Relevance Vector Machines,” Journal of Artificial Intelligence Tools (JAIT), 23 (4), 2014, pp. 1–21.
Nikolaos Kofinas, Emmanouil Orfanoudakis, Michail G. Lagoudakis, “Complete Analytical Forward and Inverse Kinematics for the NAO Humanoid Robot”, Journal of Intelligent and Robotic Systems, 2014.
Nikolaos Kofinas, Emmanouil Orfanoudakis, Michail G. Lagoudakis, “Complete Analytical Inverse Kinematics for NAO”, Proceedings of the 13th International Conference on Autonomous Robot Systems and Competitions (ROBOTICA), Lisbon, Portugal, April 2013, pp. 1–6. [Best Paper Award]
Ioannis Skoulakis and Michail G. Lagoudakis, “Efficient Reinforcement Learning in Adversarial Games,” Proceedings of the 2012 IEEE International Conference on Tools with Artificial Intelligence (ICTAI), Athens, Greece, November 2012, pp. 704–711.
Ioannis Rexakis and Michail G. Lagoudakis, “Directed Policy Search using Relevance Vector Machines,” Proceedings of the 2012 IEEE International Conference on Tools with Artificial Intelligence (ICTAI), Athens, Greece, November 2012, pp. 25–32. [Best Student Paper Award]
Alexandros Paraschos, Nikolaos Spanoudakis, and Michail G. Lagoudakis, “Model-Driven Behavior Specification for Robotic Teams,” Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Valencia, Spain, June 2012, pp. 171–178.
Vasileios Vasilikos and Michail G. Lagoudakis, “Optimization of Heuristic Search using Recursive Algorithm Selection and Reinforcement Learning,” Annals of Mathematics and Artificial Intelligence, 60 (1–2), 2010, pp. 119–151.
Jason Pazis and Michail G. Lagoudakis, “Binary Action Search for Learning Continuous-Action Control Policies,” Proceedings of the 26th International Conference on Machine Learning (ICML), Montreal, Quebec, Canada, June 2009, pp. 793–800.
Christos Dimitrakakis and Michail G. Lagoudakis, “Rollout Sampling Approximate Policy Iteration,” Machine Learning, 72 (3), 2008, pp. 157–171.
Michail G. Lagoudakis and Ronald Parr, “Least–Squares Policy Iteration,” Journal of Machine Learning Research (JMLR), 4, 2003, pp. 1107–1149.
Selected Research Projects
Project “TSI-Toshiba II: Basic Research in Statistical Spoken Dialogue Management”. This was a collaborative project between TUC-TSI and Toshiba Research Europe, Cambridge, UK to investigate integration of reinforcement learning within dialogue management. Budget: 50,000£. Duration: Feb 2016 – Feb 2018.
Project “NOPTILUS” (“autoNomous, self-Learning, OPTImal and compLete Underwater Systems”), EU FP7 ICT-funded 4-year project. NOPTILUS main objective was to determine the trajectories and the behavior of a team of Autonomous Underwater Vehicles (AUVs) that maximize situation awareness and lead to completion of their mission subject to the severe communication, sensing and environmental limitations underwater. Budget: 4,970,000€ (TUC: 524,300€). Duration: Apr 2011 – Apr 2015.
Project “RLvSL” (Reinforcement Learning via Supervised Learning). RLvSL was an EU-funded Marie Curie International Reintegration Grant 2-year project aiming to investigate the potential of using Supervised Learning to improve Reinforcement Learning and delivered algorithmic techniques which significantly improved the state-of-the-art in terms of sample efficiency and learning effectiveness. Budget: 80,000€. Duration: Dec 2006 – Dec 2008. Website: http://www.lagoudakis.gr/RLvSL
Selected Honors and Awards
2014 Kouretes video-featured on the Academic and Scientific Excellence Node, Ministry of Education
2013 Best Paper Award for article co-authored with M.Sc. student N. Kofinas, ROBOTICA 2013
2012 Best Student Paper Award for article co-authored with Ph.D. student I. Rexakis, ICTAI 2012
2011 2nd place for Noxious-Kouretes, Standard Platform League Open Challenge, RoboCup 2011
2008 3rd place for Kouretes, Standard Platform League (Nao robots), RoboCup 2008
2008 1st place for Kouretes, Microsoft Robotics Simulation Challenge, RoboCup 2008
2003 Outstanding Ph.D. Dissertation Award, Department of Computer Science, Duke University