×

zbMATH — the first resource for mathematics

Learning action models with minimal observability. (English) Zbl 07099225
Summary: This paper presents FAMA, a novel approach for learning Strips action models from observations of plan executions that compiles the learning task into a classical planning task. Unlike all existing learning systems, FAMA is able to learn when the actions of the plan executions are partially or totally unobservable and information on intermediate states is partially provided. This flexibility makes FAMA an ideal learning approach in domains where only sensoring data are accessible. Additionally, we leverage the compilation scheme and extend it to come up with an evaluation method that allows us to assess the quality of a learned model syntactically, that is, with respect to the actual model; and, semantically, that is, with respect to a set of observations of plan executions. We also show that the extended compilation scheme can be used to lay the foundations of a framework for action model comparison. FAMA is exhaustively evaluated over a wide range of IPC domains and its performance is compared to ARMS, a state-of-the-art benchmark in action model learning.
MSC:
68T Artificial intelligence
Software:
PDDL
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Kambhampati, S., Model-lite planning for the web age masses: the challenges of planning with incomplete and evolving domain models, (National Conference on Artificial Intelligence, (AAAI-07), (2007))
[2] Yang, Q.; Wu, K.; Jiang, Y., Learning action models from plan examples using weighted MAX-SAT, Artif. Intell., 171, 2-3, 107-143, (2007) · Zbl 1168.68555
[3] Amir, E.; Chang, A., Learning partially observable deterministic action models, J. Artif. Intell. Res., 33, 349-402, (2008) · Zbl 1183.68565
[4] Zhuo, H. H.; Yang, Q.; Hu, D. H.; Li, L., Learning complex action models with quantifiers and logical implications, Artif. Intell., 174, 18, 1540-1569, (2010)
[5] Zhuo, H. H.; Kambhampati, S., Action-model acquisition from noisy plan traces, (International Joint Conference on Artificial Intelligence, IJCAI-13, (2013)), 2444-2450
[6] Mourão, K.; Zettlemoyer, L. S.; Petrick, R. P.A.; Steedman, M., Learning STRIPS operators from noisy and incomplete observations, (Conference on Uncertainty in Artificial Intelligence, UAI-12, (2012)), 614-623
[7] Cresswell, S. N.; McCluskey, T. L.; West, M. M., Acquiring planning domain models using LOCM, Knowl. Eng. Rev., 28, 02, 195-213, (2013)
[8] Zhuo, H. H.; Nguyen, T. A.; Kambhampati, S., Refining incomplete planning domain models through plan traces, (International Joint Conference on Artificial Intelligence, IJCAI-13, (2013)), 2451-2458
[9] Zhuo, H. H.; Kambhampati, S., Model-lite planning: case-based vs. model-based approaches, Artif. Intell., 246, 1-21, (2017) · Zbl 1419.68102
[10] Ramírez, M.; Geffner, H., Plan recognition as planning, (International Joint conference on Artificial Intelligence, (IJCAI-09), (2009), AAAI Press), 1778-1783
[11] Ramírez, M., Plan Recognition as Planning, (2012), Universitat Pompeu Fabra, Ph.D. thesis
[12] Sohrabi, S.; Riabov, A. V.; Udrea, O., Plan recognition as planning revisited, (International Joint Conference on Artificial Intelligence, (IJCAI-16), (2016)), 3258-3264
[13] MacNally, A. M.; Lipovetzky, N.; Ramírez, M.; Pearce, A. R., Action selection for transparent planning, (International Conference on Autonomous Agents and MultiAgent Systems, (AAMAS-18), (2018)), 1327-1335
[14] Masters, P.; Sardiña, S., Deceptive path-planning, (International Joint Conference on Artificial Intelligence, (IJCAI-17), (2017)), 4368-4375
[15] Fox, M.; Long, D.; Magazzeni, D., Explainable planning, (First Workshop on Explainable AI (XAI), IJCAI-13, (2017))
[16] Chakraborti, T.; Sreedharan, S.; Kambhampati, S., Explicability versus explanations in human-aware planning, (International Conference on Autonomous Agents and MultiAgent Systems, AAMAS-18, (2018)), 2180-2182
[17] Pozanco, A.; E.-Martín, Y.; Fernández, S.; Borrajo, D., Counterplanning using goal recognition and landmarks, (International Joint Conference on Artificial Intelligence, (IJCAI-18), (2018)), 4808-4814
[18] Chakraborti, T.; Sreedharan, S.; Zhang, Y.; Kambhampati, S., Plan explanations as model reconciliation: moving beyond explanation as soliloquy, (International Joint Conference on Artificial Intelligence, (IJCAI-17), (2017)), 156-163
[19] Bonet, B.; Palacios, H.; Geffner, H., Automatic Derivation of Memoryless Policies and Finite-State Controllers Using Classical Planners, (International Conference on Automated Planning and Scheduling, (ICAPS-09), (2009), AAAI Press)
[20] Aguas, J. S.; Celorrio, S. J.; Jonsson, A., Generalized planning with procedural domain control knowledge, (International Conference on Automated Planning and Scheduling, (ICAPS-16), (2016), AAAI Press), 285-293
[21] Segovia-Aguas, J.; Jiménez, S.; Jonsson, A., Hierarchical finite state controllers for generalized planning, (International Joint Conference on Artificial Intelligence, (IJCAI-16), (2016), AAAI Press), 3235-3241
[22] Segovia-Aguas, J.; Jiménez, S.; Jonsson, A., Generating context-free grammars using classical planning, (International Joint Conference on Artificial Intelligence, (IJCAI-17), (2017), AAAI Press), 4391-4397
[23] Fikes, R. E.; Nilsson, N. J., Strips: a new approach to the application of theorem proving to problem solving, Artif. Intell., 2, 3-4, 189-208, (1971) · Zbl 0234.68036
[24] Konidaris, G.; Kaelbling, L. P.; Lozano-Pérez, T., From skills to symbols: learning symbolic representations for abstract high-level planning, J. Artif. Intell. Res., 61, 215-289, (2018) · Zbl 1426.68254
[25] Asai, M.; Fukunaga, A., Classical planning in deep latent space: bridging the subsymbolic-symbolic boundary, (National Conference on Artificial Intelligence, AAAI-18, (2018))
[26] Aineto, D.; Jiménez, S.; Onaindia, E., Learning STRIPS action models with classical planning, (International Conference on Automated Planning and Scheduling, (ICAPS-18), (2018), AAAI Press), 399-407
[27] Zhuo, H. H., Crowdsourced action-model acquisition for planning, (National Conference on Artificial Intelligence, AAAI-15, (2015)), 3439-3446
[28] Davis, J.; Goadrich, M., The relationship between precision-recall and ROC curves, (International Conference on Machine Learning, (2006), ACM), 233-240
[29] Ginsberg, M. L.; Smith, D. E., Reasoning about action II: the qualification problem, Artif. Intell., 35, 3, 311-342, (1988) · Zbl 0645.68110
[30] Kulkarni, A.; Chakraborti, T.; Zha, Y.; Vadlamudi, S. G.; Zhang, Y.; Kambhampati, S., Explicable robot planning as minimizing distance from expected behavior, CoRR
[31] Fox, M.; Long, D., PDDL2.1: an extension to PDDL for expressing temporal planning domains, J. Artif. Intell. Res., 20, 61-124, (2003) · Zbl 1036.68093
[32] Cresswell, S.; McCluskey, T. L.; West, M. M., Acquisition of object-centred domain models from planning examples, (International Conference on Automated Planning and Scheduling, (ICAPS-09), (2009), AAAI Press)
[33] Cresswell, S.; Gregory, P., Generalised domain model acquisition from action traces, (International Conference on Automated Planning and Scheduling, ICAPS-11, (2011), AAAI Press)
[34] Gregory, P.; Cresswell, S., Domain model acquisition in the presence of static relations in the LOP system, (International Joint Conference on Artificial Intelligence, (IJCAI-16), (2016)), 4160-4164
[35] McDermott, D.; Ghallab, M.; Howe, A.; Knoblock, C.; Ram, A.; Veloso, M.; Weld, D.; Wilkins, D., PDDL - The Planning Domain Definition Language, (1998)
[36] Slaney, J.; Thiébaux, S., Blocks world revisited, Artif. Intell., 125, 1-2, 119-153, (2001) · Zbl 0969.68136
[37] Cook, S. A., The complexity of theorem-proving procedures, (Proceedings of the Third Annual ACM Symposium on Theory of Computing, (1971), ACM), 151-158
[38] Jiménez, S.; De La Rosa, T.; Fernández, S.; Fernández, F.; Borrajo, D., A review of machine learning for automated planning, Knowl. Eng. Rev., 27, 04, 433-467, (2012)
[39] Stern, R.; Juba, B., Efficient, safe, and probably approximately complete learning of action models, (International Joint Conference on Artificial Intelligence, (IJCAI-17), (2017)), 4405-4411
[40] Kucera, J.; Barták, R., LOUGA: learning planning operators using genetic algorithms, (Pacific Rim Knowledge Acquisition Workshop, PKAW-18, (2018)), 124-138
[41] Howey, R.; Long, D.; Fox, M., VAL: automatic plan validation, continuous effects and mixed initiative planning using PDDL, (Tools with Artificial Intelligence, 2004. ICTAI 2004. 16th IEEE International Conference on, (2004), IEEE), 294-301
[42] C. Muise, Planning.domains, ICAPS system demonstration.
[43] Rintanen, J., Madagascar: scalable planning with SAT, (International Planning Competition, (IPC-2014), (2014))
[44] López, C. L.; Celorrio, S. J.; Olaya, Á. G., The deterministic part of the seventh international planning competition, Artif. Intell., 223, 82-119, (2015)
[45] Fern, A.; Yoon, S. W.; Givan, R., Learning domain-specific control knowledge from random walks, (International Conference on Automated Planning and Scheduling, ICAPS-04, (2004), AAAI Press), 191-199
[46] Francès, G.; Ramírez, M.; Lipovetzky, N.; Geffner, H., Purely declarative action descriptions are overrated: classical planning with simulators, (International Joint Conference on Artificial Intelligence, (IJCAI-17), (2017)), 4294-4301
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.