Hector Geffner

I am a researcher at the Institució Catalana de Recerca i Estudis Avançats (ICREA) and a professor at the  Departament de Tecnologies de la Informació i les Comunicacions (DTIC),  Universitat Pompeu Fabra, in the Artificial Intelligence Group. I'm also a Wallenberg Guest Professor at Linköping University, Sweden.

My address is:

Hector Geffner
DTIC, Universitat Pompeu Fabra
C/Roc Boronat 138, Office 55.213
E-08018 Barcelona, Spain

+34 93 542 2563 (voice)
+34 93 542 2451  (fax)

firstname.lastname@upf.edu

Index

• Software
• Opportunities
• Papers
• Short Bio

Software 

• Software and example files for each of the papers should be available from co-author pages; if not, feel free to request them by email
  

Opportunities

• We have currently a number of funded slots for PhD students and Postdocs in the context of the project  Representation Learning for Planning (RLeap), funded by an Advanced ERC Grant (2020-2025)
  

Papers, Talks, Slides, Books

• Learning Generalized Policies Without Supervision Using GNNs. S Stählberg, B Bonet, H Geffner. KR 2022
• Learning First-Order Symbolic Planning Representations That Are Grounded. A Occhipinti, B Bonet, H Geffner. ICAPS PRL Workshop, 2022.
• Target Languages (vs. Inductive Biases) for Learning to Act and Plan. H Geffner. AAAI 2022 (Slides)
• Learning Sketches for Decomposing Planning Problems into Subproblems of Bounded Width. D Drexler, J Seipp, H Geffner. ICAPS 2022
• Learning General Optimal Policies with Graph Neural Networks. S Stählberg, B Bonet, H Geffner. ICAPS 2022
• Probabilistic and Causal Inference: The Works of Judea Pearl. H. Geffner. R. Dechter. Joseph Y. Halpern (Eds). ACM Books, 2022.
• Video: Target Languages for Learning to Act & Plan. H. Geffner. IJCLR 2021 (Invited). (Slides)
• Learning First-Order Representations for Planning from Black-Box States: New Results. Iván D. Rodriguez, B Bonet, J Romero, H Geffner. KR 2021
• Expressing and Exploiting the Common Subgoal Structure of Classical Planning Domains Using Sketches.  D Drexler, J Seipp, H Geffner. KR 2021
  
• Flexible FOND Planning with Explicit Fairness Assumptions. Iván D. Rodriguez, B Bonet, S Sardiña, H Geffner. ICAPS 2021.
  
• General Policies, Representations, and Planning Width. B Bonet, H Geffner.  AAAI 2021.
• Learning General Planning Policies from Small Examples Without Supervision. G Francès, B Bonet, H Geffner. AAAI 2021.
  
• Video: Model-free, model-based, and general intelligence. H. Geffner. MIT Embodied Intelligence Seminar, 4/2020.
  
• High-Level Programming via Generalized Planning and LTL Synthesis. B Bonet, G De Giacomo, H Geffner, F Patrizi, S Rubin. KR 2020 
• Learning First-Order Symbolic Representations for Planning from the Structure of the State Space, B. Bonet, H. Geffner. ECAI-2020
• Qualitative Numeric Planning: Reductions and Complexity B Blai, H Geffner. Journal of Artificial Intelligence Research, 2020
  
• From data-based to model-based AI: Representation learning for planning (RLeap). H. Geffner.  Edited ERC Proposal, 8/2019. (Short Version).
• Learning Features and Abstract Actions for Computing Generalized Plans, B. Bonet, G. Frances, H. Geffner. Proc. AAAI-2019
  
• Features, Projections, and Representation Change for Generalized Planning. B. Bonet, H. Geffner. Proc. IJCAI 2018
  
• Compact Policies for Fully-Observable Non-Deterministic Planning as SAT. T. Geffner, H. Geffner. Proc. ICAPS 2018 (Software)
  
• Planning with pixels in (almost) real time.W. Bandres, B. Bonet, H. Geffner.   Proc. AAAI 2018
• Video: Model-free, Model-based, and General Intelligence. H. Geffner. Proc. IJCAI 2018 (Invited). Slides. Video.
• Purely Declarative Action Descriptions are Overrated: Classical Planning with Simulators. G. Frances, M. Ramirez, N. Lipovetzky, H. Geffner. Proc. IJCAI 2017
• Generalized Planning: Non-Deterministic Abstractions and Trajectory Constraints. B. Bonet, G. De Giacomo, H. Geffner, S. Rubin. Proc. IJCAI 2017
  
• A Polynomial Planning Algorithm that Beats LAMA and FF. N Lipovetzky, H Geffner. Proc. ICAPS 2017
• Multiagent Online Planning with Nested Beliefs and Dialogue. F. Kominis,and H. Geffner, Proc. ICAPS 2017.
• Best-First Width Search: Exploration and Exploitation in Classical Planning. N Lipovetzky, H Geffner. Proc. AAAI 2017
• Artificial Intelligence and the Common Good (Slides). H. Geffner. B-Debate AI: Dreams, Risks, and Reality. CosmoCaixa, Barcelona 2017
  
• Combined Task and Motion Planning as Classical AI Planning. J. Ferrer-Mestres, G. Frances, H. Geffner. 2016
  
• Traps, invariants, and dead-ends. N Lipovetzky, C Muise, H Geffner. Proc. ICAPS 2016
  
• E-STRIPS: Existential Quantification in Planning and Constraint Satisfaction. G Frances, H Geffner. Proc. IJCAI 2016
  
• Effective Planning with More Expressive Languages. G Frances, H Geffner. Proc. IJCAI 2016
  
• Factored Probabilistic Belief Tracking. B Bonet, H Geffner. Proc. IJCAI 2016
  
• Video: General Solvers for General Artificial Intelligence. H. Geffner, UPF 6/2016
  
• Planning with State Constraints and its Application to Combined Task and Motion Planning. J. Ferrer-Mestres, G. Frances, H. Geffner 2015 ICAPS PlanRob Workshop
  
• Policies that Generalize: Solving Many planning Problems with the Same Policy. B. Bonet and H. Geffner. Proc. IJCAI-2015
  
• Classical Planning with Simulators: Results on the Atari Video Games. N. Lipovetzky, M. Ramirez, H. Geffner. Proc. IJCAI-2015
  
• Modeling and Computation in Planning: Better Heuristics for More Expressive Languages. G. Frances and H. Geffner. Proc. ICAPS-2015
• Beliefs in Multiagent Planning: From One Agent to Many. F. Kominis and H. Geffner. Proc. ICAPS-2015
• Video: Present and Future of Artificial Intelligence (in Spanish). H. Geffner, CCCB 12/2015
• Belief Tracking for Planning with Sensing: Width, Complexity, and Approximations. B. Bonet and H. Geffner. JAIR 2014.
  
• Width-based Algorithms for Classical Planning: New Results. N. Lipovetzky and H. Geffner. Proc. ECAI-2014
  
• Flexible and Scalable Partially Observable Planning with Linear Translations. B. Bonet and H. Geffner. Proc. AAAI-2014
• Non-classical Planning wth a Classical Planner: The Power of Transformations. H. Geffner. Proc. JELIA-2014. Springer.
• Artificial Intelligence: From Programs to Solvers. H. Geffner. AI Communications. 2014 
  
• Book: A Concise Introduction to Models and Methods for Automated Planning. H. Geffner and B. Bonet. Morgan & Claypool, 2013 (fragment, slides).
  
• Computational Models of Planning. H. Geffner.  Wiley Interdisciplinary Reviews: Cognitive Science, 2013
  
• Causal Belief Decomposition for Planning with Sensing: Completeness Results and Practical Approximation. B. Bonet and H. Geffner. Proc. IJCAI-2013, Beijing.
  
• Fair LTL Synthesis for Non-Deterministic Systems using Strong Cyclic Planners. F. Patrizi, N. Lipovetzky, H. Geffner. Proc. IJCAI-2013, Beijing.
  
• Width and Serialization of Classical Planning Problems. N. Lipovetzky and H. Geffner. Proc. ECAI-2012, Montpellier. (Slides)
  
• Action Selection for MDPs: Anytime AO* vs. UCT. B. Bonet and H. Geffner. Proc. AAAI-2012, Toronto.
  
• Width and Complexity of Belief Tracking in Non-Deterministic Conformant and Contingent Planning. B. Bonet and H. Geffner. Proc. AAAI-2012.
  
• Planning under Partial Observability by Classical Replanning: Theory and Experiments. B. Bonet and H. Geffner. Proc. IJCAI-2011, Barcelona
  
• Goal Recognition over POMDPs: Inferring the Intention of a POMDP Agent. M. Ramirez and H. Geffner. Proc. IJCAI-2011, Barcelona
  
• Computing Infinite Plans for LTL Goals using a Classical Planner. F. Patrizi, N. Lipovetzky, G. De Giacomo, H. Geffner. Proc IJCAI-2011, Barcelona
• Advanced Introduction to Planning: Models and Methods, H. Geffner. Tutorial at IJCAI-2011, Barcelona
  
• Qualitative Numeric Planning. S. Srivastava, S. Zilberstein, N. Immerman, H. Geffner. Proc. AAAI-2011, San Francisco
  
• Effective Heuristics and Belief Tracking for Planning with Incomplete Information. A. Albore, M. Ramirez, H. Geffner. Proc. ICAPS-2011, Freiburg
• Searching for Plans with Carefully Designed Probes. N. Lipovetzky and H. Geffner. Proc. ICAPS-2011, Freiburg
• Heuristic Search for Generalized Stochastic Shortest Path MDPs. A. Kolobov, Mausam, D. Weld, H. Geffner, Proc. ICAPS-2011, Freiburg
• Planning and Plan Recognition (Slides), Invited talk, Dagstuhl Plan Recognition Seminar, 2011
  
• Heuristics with Choice Variables: Bridging the Gap between Planning and Graphical Models, E. Keyder, M. Ramirez, H. Geffner. ICAPS Workshop  2011
• Course on Automated Planning (Slides). Univ. La Sapienza, Rome, Summer 2010
• The Model-based Approach to Autonomous Behavior: Prospects and Challenges (Slides), Invited talk, AI*IA, Brescia, 12/2010
  
• Heuristics, Probability and Causality: A Tribute to Judea Pearl. R. Dechter, H. Geffner, J. Halpern (Eds). College Publications, 2010 (Table of Contents)
• Heuristics, Planning, and Cognition. In Heuristics, Probability and Causality: A Tribute to Judea Pearl. R. Dechter, H. Geffner, J. Halpern (Eds), College Publications, 2010.
• Probabilistic Plan Recognition using off-the-shelf Classical Planners. M. Ramirez and H. Geffner. Proc. AAAI-10. Atlanta, USA. 7/2010
• The Model-based Approach to Autonomous Behavior: A Personal View. H. Geffner. Proc. AAAI-10. Atlanta, USA. 7/2010
• Automatic Derivation of Finite-State Machines for Behavior Control. B. Bonet, H. Palacios. H. Geffner. Proc. AAAI-10. Atlanta, USA. 7/2010
• Compiling Away Uncertainty in Non-Deterministic Conformant Planning Problems. A. Albore, H. Palacios, H. Geffner. Proc. ECAI-10, Lisbon, Portugal, 8/2010
• Soft Goals Can Be Compiled Away. E. Keyder and H. Geffner. Journal of Artificial Intelligence Research, Volume 36, pages 547-556, 2009.
• Inference and Learning in Planning (Extended Abstract). H. Geffner. Proc. Discovery Science 2009, Porto, Portugal, 10/2009.
• Automatic Derivation of Memoryless Policies and Finite-State Controllers Using Classical Planners. B. Bonet, H. Palacios, H. Geffner. Proc. ICAPS-09. 
• Inference and Decomposition in Planning Using Causal Consistent Chains. N. Lipovetzky and H. Geffner. Proc. ICAPS-09. 
• Plan Recognition as Planning. M. Ramirez and H. Geffner. Proc. IJCAI-09. Pasadena, USA. 7/09.
• A Translation-based Approach to Contingent Planning. A. Albore, H. Palacios, H. Geffner. Proc. IJCAI-09. Pasadena, USA. 7/09.
• Trees of Shortest Paths vs. Steiner Trees: Understanding and Improving Delete Relaxation Heuristics. E. Keyder and H. Geffner. Proc. IJCAI-09. 
• Solving POMDPs: RTDP-Bel vs. Point-based Algorithms. B. Bonet and H. Geffner. Proc. IJCAI-09. 
• Compiling Uncertainty Away in Conformant Planning Problems with Bounded Width. H. Palacios and H. Geffner Journal of AI Research (JAIR). 2009.
• Heuristics for Planning with Penalties and Rewards Formulated in Logic and Computed Through Circuits, B. Bonet and H. Geffner. Artificial Intelligence, 2008 (172) 1579-1604
• Heuristics for Planning with Action Costs Revisited. E. Keyder and H. Geffner. Proc. ECAI-08.
• Unifying the Causal Graph and Additive Heuristics, M. Helmert and H. Geffner, Proc. ICAPS 2008.
  
• AI at 50: From Programs to Solvers. Models and Techniques for General Intelligence (Slides). H. Geffner. University of Edinburgh, School of Informatics. Distinguished Lecture Series. 12/2007
• From Conformant into Classical Planning: Efficient Translations that may be Complete Too. H. Palacios and H. Geffner. Proc. ICAPS-2007
  
• Structural Relaxations by Variable Renaming and their Compilation for  Solving MinCostSAT. M. Ramirez and H. Geffner. Proc. CP-2007
  
• The Causal Graph Heuristic is the Additive Heuristic plus Context. H. Geffner. Proc. 2007 ICAPS HSDIP Workshop 
  
• Fast and Informed Action Selection for Planning with Sensing. A. Albore, H. Palacios, H. Geffner. Proc.  CAEPIA-07, Springer
• Set-Additive and TSP Heuristics for Planning with Action Costs and Soft Goals. E. Keyder and H. Geffner. Proc. 2007 HSDIP ICAPS 
  
• Logical Encodings with No Time Indexes for Defining and Computing Admissible Heuristics for Planning. M. Ramirez, B. Bonet, H. Geffner. Proc. 2007 ICAPS Workshop.
  
• Heuristics for Planning with Action Costs. E. Keyder and H. Geffner. Proc.  12th Conf.  Spanish AI (CAEPIA-07),  Springer.
  
• Compiling Uncertainty Away: Solving Conformant Planning Problems Using a Classical Planner (Sometimes). H. Palacios and H. Geffner. Proc.  AAAI-2006
• Heuristics for Planning with Penalties and Rewards using Compiled Knowledge. B. Bonet and H. Geffner. Proc. KR-2006
  
• Learning Depth-First Search: A Unified Approach to Heuristic Search in Deterministic and Non-Deterministic Settings, and its application to MDPs. B. Bonet and H. Geffner. Proc. ICAPS-2006
  
• Branching and Pruning: An Optimal Temporal POCL Planner based on Constraint Programming (Extended Version). V. Vidal and H. Geffner. Artificial Intelligence,  Vol 170(3), pp  298-335,  March 2006.
• mGPT: a Probabilistic Planner based on Heuristic Search. B. Bonet and H. Geffner. Journal of AI Research, Vol 24, pp 933-944, 2005.
• Search and Inference in AI Planning. H. Geffner. Proc. CP-2005 Invited talk (Slides)
• Mapping Conformant Planning into SAT through Compilation and Projection. H. Palacios and H. Geffner.  Lecture Notes in Computer Sciences, vol 4177.  Proc. CAEPIA-05  (originally in  Spanish)
• Learning in Depth-First Search: A Unified Approach to Heuristic Search in Deterministic, Non-Deterministic, Probabilistic, and Game Tree Settings. B. Bonet and H. Geffner. 2005.
• Solving Simple Planning Problems with More Inference and No Search, V. Vidal and H. Geffner. Proc. CP-2005.
• Selecting Actions and Making Decisions: Lessons from AI Planning. H. Geffner. Proc. IJCAI-05 MNAS-2005 Workshop on Modeling Natural Action Selection (MNAS-05).  (Slides)
• An Algorithm Better than AO*? B. Bonet and H. Geffner. Proc. AAAI-2005 (Slides)
• New Admissible Heuristics for Domain-Independent Planning. P. Haslum, B. Bonet and H. Geffner. Proc. AAAI-05.
  
• Pruning Conformant Plans by Counting Models on Compiled d-DNNF Representations.H. Palacios, B. Bonet, A. Darwiche, H. Geffner. Proc. ICAPS-2005.
• Learning Generalized Policies from Planning Examples Using Concept Languages (Extended and Revised Version).  M. Martin and H. Geffner, Applied Intelligence 20 (1) , 9-19, 2004
• Branching and Pruning: An Optimal Temporal POCL Planner based on Constraint Programming. V. Vidal and H. Geffner.  Proc. AAAI-04
  
• Planning Graphs and Knowledge Compilation. H. Geffner. Proc. KR-04.  (Slides)
• PDDL 2.1: Representation vs. Computation.H. Geffner (a commentary on PDDL 2.1 and the 3rd Int. Planning Competition). J. of AI Research, vol 20, 2003.
• Faster Heuristic Search Algorithms for Planning with Uncertainty and Full Feedback. B. Bonet and H. Geffner. Proc. IJCAI 2003
  
• Labeled RTDP: Improving the Convergence of Real Time Dynamic Programming. B. Bonet and H. Geffner. Proc. ICAPS-2003
  
• Branching Matters: Alternative Branching in Graphplan. J. Hoffmann and H. Geffner. Proc. ICAPS-2003
  
• Perspectives on Artificial Intelligence Planning. H. Geffner Proc. AAAI-2002.  Invited talk (Slides)
• Planning as Branch and Bound: A Constraint Programming Implementation;  H. Palacios and H. Geffner. XVIII Latin-American Conf. on Informatics (CLEI-2002)
• Heuristic Search Planning: Progress and Challenges , Slides Invited Talk at European Conf. on Planning (ECP-2001),  (4 slides per page)
• Planning as Branch and Bound and its Relation to Constraint-based Approaches;  H. Geffner, 4/2001
• GPT: A Tool for Planning with Uncertainty and Partial Information;  B. Bonet and H. Geffner;  IJCAI-2001 Workshop on Planning with Incomplete Info, 8/2001
• Planning and Control in Artificial Intelligence: a Unifying Perspective.  B. Bonet and H. Geffner. Applied Intelligence, Vol 14(3), 2001, pp  237--252.
• Heuristic Planning with Time and Resources;  P. Haslum and H. Geffner,  Proceedings ECP-2001  4/2001, Springer
• HSP2:  Description HSP Planner  in AIPS-2000 Competition, B. Bonet and H. Geffner, AI Magazine Vol 22 (3), 2001, pp 77--80
• Planning as Heuristic Search  B. Bonet and H. Geffner.  Artificial Intelligence, Special issue on Heuristic Search. Vol 129 (1-2) 2001
• Planning with Incomplete Information as Heuristic Search in Belief Space B. Bonet and H. Geffner. Proc. AIPS-2000 
• Admissible Heuristics for Optimal Planning P. Haslum H. Geffner. Proc. AIPS-2000 
• Learning generalized policies in planning using concept languages M. Martin and H. Geffner. Proc. KR-2000 (Slides)
• HSP: Heuristic Search Planner B. Bonet and H. Geffner. Entry at AIPS-98 Planning Competition, AI Magazine Vol 21(2), 2000
• Functional Strips: a more flexible language for planning and problem solving H. Geffner. In Logic-Based Artificial Intelligence, Jack Minker (Ed.), Kluwer, 2000.
• Planning as Heuristic Search: New Results B. Bonet and H. Geffner. Proc. European Conference on Planning (ECP-99), Springer. (Slides)
• Modelling Intelligent Behaviour: the Markov Decision Approach H. Geffner. Invited talk IBERAMIA 1998, Lect. Notes in AI 1484, H. Coelho (Ed), Springer,  
• Classical, Probabilistic and Contingent Planning: Three Models, One Algorithm H. Geffner. Proceedings AIPS'98 Workshop on Planning as Combinatorial Search. (Slides)
• High-Level Planning and Control with Incomplete Information Using POMDPs H. Geffner and B. Bonet. Proceedings Fall AAAI Symposium on Cognitive Robotics, 1998. (Slides)
• Solving Large POMDPs by Real Time Dynamic Programming H. Geffner and B. Bonet. Working Notes Fall AAAI Symposium on POMDPS. 1998.
• Modeling action, knowledge and control H. Geffner and J. Wainer. Proceedings European Conference on Artificial Intelligence (ECAI-98), Brighton, UK. 8/98.
• Learning Sorting and Decision Trees with POMDPs B. Bonet and H. Geffner. Proc ICML-98. (Slides)
• Causality, Constraints and the Side Effects of Actions H. Geffner. Proceedings IJCAI-97, Naogoya, Japan, 8/97.
• A fast and robust action selection mechanism for planning B. Bonet, G. Loerincs, H. Geffner. Proceedings AAAI-97. Provide, RI, 7/97. (Slides)
• An Analysis of control programs using models of action. J. Ramirez and H. Geffner. AAAI-97 Workshop on Action. 1997 (Slides) 

Short Bio

Hector Geffner got his Ph.D at UCLA in 1989. He then worked as Staff Research Member at the IBM T.J. Watson Research Center in NY, USA and at the Universidad Simon Bolivar, in Caracas, Venezuela. Since 2001, he is a researcher at ICREA and a professor at the Universitat Pompeu Fabra, Barcelona, and since 2019, a Wallenberg Guest Professor at Linköping University, Sweden. Hector is a former Associate Editor of Artificial Intelligence and the Journal of Artificial Intelligence Research, and a Fellow of AAAI and EurAI. He is the author of the book Default Reasoning: Causal and Conditional Theories, MIT Press, 1992, and "A Concise Introduction to Models and Methods for Automated Planning" with Blai Bonet, Morgan and Claypool, 2013. He edited two books in tribute of Judea Pearl with Rina Dechter and Joe Halpern: "Heuristics, Probability, and Causality: a Tribute to Judea Pearl", College Publications, 2010, and Probabilistic and Causal Inference: The Works of Judea Pearl, ACM Books 2022. Hector is interested in computational models of reasoning, action, learning, and planning that are general and effective. He is also a concerned citizen (particularly concerned these days) and, aside from courses on logic and AI, he teaches a course on social and technological change. He leads a project on representation learning for acting and planning (RLeap), funded by an Advanced ERC grant, 2020-2025.