{"type":"ScholarlyArticle","title":"Learning Neural-Symbolic Descriptive Planning Models via Cube-Space Priors: The Voyage Home (to STRIPS)","authors":[{"name":"Masataro Asai","slug":"masataro-asai","given":"Masataro","family":"Asai","affiliation":"MIT-IBM Watson AI Lab , IBM Research , Cambridge USA"},{"name":"Christian Muise","slug":"christian-muise","given":"Christian","family":"Muise","affiliation":"School of Computing , Queen's University , Kingston Canada"}],"abstract":"We achieved a new milestone in the difficult task of enabling agents to learn about their environment autonomously. Our neuro-symbolic architecture is trained end-to-end to produce a succinct and effective discrete state transition model from images alone. Our target representation (the Planning Domain Definition Language) is already in a form that off-the-shelf solvers can consume, and opens the door to the rich array of modern heuristic search capabilities. We demonstrate how the sophisticated innate prior we place on the learning process significantly reduces the complexity of the learned representation, and reveals a connection to the graph-theoretic notion of ``cube-like graphs'', thus opening the door to a deeper understanding of the ideal properties for learned symbolic representations. We show that the powerful domain-independent heuristics allow our system to solve visual 15-Puzzle instances which are beyond the reach of blind search, without resorting to the Reinforcement Learning approach that requires a huge amount of training on the domain-dependent reward information.","doi":"10.24963/ijcai.2020/371","url":"https://ijcai.org/proceedings/2020/371","pdf":"https://ijcai.org/proceedings/2020/0371.pdf","markdown":"https://ijcai.org/proceedings/2020/371.md","bibtex":"https://ijcai.org/proceedings/2020/bibtex/371","volume":{"slug":"2020","title":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2020","url":"https://ijcai.org/proceedings/2020/","year":2020,"location":"Yokohama","dates":"January 2021 (postponed from 11-17 July 2020)","editors":"Christian Bessiere","isbn":"978-0-9992411-6-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2020"},"section":"Machine Learning","track":"Main track","number":371,"pages":{"first":2676,"last":2682},"keywords":["Planning and Scheduling: Planning and Scheduling","Machine Learning: Neuro-Symbolic Methods"],"videos":[{"label":"Video","url":"https://videos.ijcai20.org/24912/file_24912.m3u8","length_sec":277,"provider":"hls"}],"references":[{"position":1,"raw":"[Aineto et al., 2018] Diego Aineto, Sergio Jiménez, and Eva Onaindia. Learning STRIPS Action Models with Classical Planning. In ICAPS, 2018.","title":"Learning STRIPS Action Models with Classical Planning","authors":"Aineto","year":"2018","venue":"ICAPS","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Amado et al., 2018] Leonardo Amado, Ramon Fraga Pereira, João Paulo Aires, Mauricio Cecilio Mag- naguagno, Roger Granada, and Felipe Meneguzzi. Goal Recognition in Latent Space. In IJCNN, 2018.","title":"Goal Recognition in Latent Space","authors":"Leonardo Amado, Ramon Fraga Pereira, João, Paulo Aires, Mauricio Cecilio Mag-Naguagno, Roger Granada, Felipe Meneguzzi","year":"2018","venue":"IJCNN","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Asai and Fukunaga, 2018] Masataro Asai and Alex Fuku- naga. Classical Planning in Deep Latent Space: Bridging the Subsymbolic-Symbolic Boundary. In AAAI, 2018.","title":"Masataro Asai and Alex Fukunaga. Classical Planning in Deep Latent Space: Bridging the Subsymbolic-Symbolic Boundary","authors":"Fukunaga Asai","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Asai and Kajino, 2019] Masataro Asai and Hiroshi Kajino. Towards Stable Symbol Grounding with Zero-Suppressed State AutoEncoder. In ICAPS, July 2019.","title":"Towards Stable Symbol Grounding with Zero-Suppressed State AutoEncoder","authors":"Kajino Asai, Masataro Asai, Hiroshi Kajino","year":"2019","venue":"ICAPS","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Asai, 2019] Masataro Asai. Neural-Symbolic Descriptive Action Model from Images: The Search for STRIPS. arXiv:1912.05492, 2019.","title":"Masataro Asai","authors":"Asai","year":"2019","venue":"Neural-Symbolic Descriptive Action Model from Images: The Search for STRIPS","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Bengio et al., 2013] Yoshua Bengio, Nicholas Léonard, and Aaron Courville. Estimating or Propagating Gradients through Stochastic Neurons for Conditional Computation. arXiv:1308.3432, 2013.","title":"Estimating or Propagating Gradients through Stochastic Neurons for Conditional Computation","authors":"Yoshua Bengio, Nicholas Léonard, Aaron Courville","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Bonet and Geffner, 2020] Blai Bonet and Hector Geffner. Learning First-Order Symbolic Representations for Plan- ning from the Structure of the State Space. In ECAI, 2020.","title":"Learning First-Order Symbolic Representations for Planning from the Structure of the State Space","authors":"Geffner Bonet, Blai Bonet, Hector Geffner","year":"2020","venue":"ECAI","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Cresswell et al., 2013] Stephen Cresswell, Thomas Leo McCluskey, and Margaret Mary West. Acquiring planning domain models using LOCM. Knowledge Engineering Re- view, 28(2), 2013.","title":"Acquiring planning domain models using LOCM","authors":"Stephen Cresswell, Thomas Leo Mccluskey, Margaret Mary West","year":"2013","venue":"Knowledge Engineering Review","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Elkan and Noto, 2008] Charles Elkan and Keith Noto. Learning Classifiers from Only Positive and Unlabeled Data. In KDD. ACM, 2008.","title":"Learning Classifiers from Only Positive and Unlabeled Data","authors":"Noto Elkan, Charles Elkan, Keith Noto","year":"2008","venue":"KDD. ACM","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Feng et al., 2018] Wenfeng Feng, Hankz Hankui Zhuo, and Subbarao Kambhampati. Extracting Action Sequences from Texts Based on Deep Reinforcement Learning. In IJCAI, 2018.","title":"Extracting Action Sequences from Texts Based on Deep Reinforcement Learning","authors":"Wenfeng Feng, Hankz Hankui Zhuo, Subbarao Kambhampati","year":"2018","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2018/565"},{"position":11,"raw":"[Fikes and Nilsson, 1972] Richard E Fikes and Nils J. Nils- son. STRIPS: A New Approach to the Application of The- orem Proving to Problem Solving. Artificial Intelligence, 2(3), 1972.","title":"STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving","authors":"E Richard, Nils J Fikes, Nilsson","year":"1972","venue":"Artificial Intelligence","doi":null,"url":null,"internal_url":"/proceedings/1971/63"},{"position":12,"raw":"[Gebser et al., 2011] Martin Gebser, Benjamin Kaufmann, Roland Kaminski, Max Ostrowski, Torsten Schaub, and Marius Schneider. Potassco: The Potsdam answer set solv- ing collection. Ai Communications, 24(2):107-124, 2011.","title":"Potassco: The Potsdam answer set solving collection","authors":"Martin Gebser, Benjamin Kaufmann, Roland Kaminski, Max Ostrowski, Torsten Schaub, Marius Schneider","year":"2011","venue":"Ai Communications","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Hart et al., 1968] Peter E. Hart, Nils J. Nilsson, and Bertram Raphael. A Formal Basis for the Heuristic Determination of Minimum Cost Paths. Systems Science and Cybernetics, IEEE Transactions on, 4(2), 1968.","title":"A Formal Basis for the Heuristic Determination of Minimum Cost Paths","authors":"Peter E Hart, Nils J Nilsson, Bertram Raphael","year":"1968","venue":"Systems Science and Cybernetics, IEEE Transactions on","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Haslum et al., 2019] Patrik Haslum, Nir Lipovetzky, Daniele Magazzeni, and Christian Muise. An Introduction to the Planning Domain Definition Language. Synthesis Lectures on Artificial Intelligence and Machine Learning, 13(2):1-187, 2019.","title":"An Introduction to the Planning Domain Definition Language","authors":"Haslum","year":"2019","venue":"Synthesis Lectures on Artificial Intelligence and Machine Learning","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[He et al., 2016] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep Residual Learning for Image Recognition. In CVPR, 2016.","title":"Deep Residual Learning for Image Recognition","authors":"Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun","year":"2016","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Helmert and Domshlak, 2009] Malte Helmert and Carmel Domshlak. Landmarks, Critical Paths and Abstractions: What's the Difference Anyway? In ICAPS, 2009. [Helmert et al., 2014] Malte Helmert, Patrik Haslum, Jörg Hoffmann, and Raz Nissim. Merge-and-Shrink Abstrac- tion: A Method for Generating Lower Bounds in Factored State Spaces. Journal of ACM, 61(3), 2014.","title":"Merge-and-Shrink Abstraction: A Method for Generating Lower Bounds in Factored State Spaces","authors":"Domshlak Helmert, Malte Helmert, Carmel Domshlak, ; Helmert","year":"2009","venue":"ICAPS","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Higgins et al., 2017] Irina Higgins, Loïc Matthey, Arka Pal, et al. β-VAE: Learning Basic Visual Concepts with a Con- strained Variational Framework. In ICLR, 2017.","title":"β-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework","authors":"Irina Higgins, Loïc Matthey, Arka Pal","year":"2017","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Ioffe and Szegedy, 2015] Sergey Ioffe and Christian Szegedy. Batch Normalization: Accelerating Deep Net- work Training by Reducing Internal Covariate Shift. In ICML, 2015.","title":"Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift","authors":"Sergey Ioffe, Christian Szegedy","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Jang et al., 2017] Eric Jang, Shixiang Gu, and Ben Poole. Categorical Reparameterization with Gumbel-Softmax. In ICLR, 2017.","title":"Categorical Reparameterization with Gumbel-Softmax","authors":"Eric Jang, Shixiang Gu, Ben Poole","year":"2017","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Koul et al., 2019] Anurag Koul, Alan Fern, and Sam Grey- danus. Learning Finite State Representations of Recurrent Policy Networks. In ICLR, 2019.","title":"Learning Finite State Representations of Recurrent Policy Networks","authors":"Anurag Koul, Alan Fern, Sam Greydanus","year":"2019","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[LeCun et al., 1998] Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-Based Learning Applied to Document Recognition. Proc. of the IEEE, 86(11), 1998.","title":"Gradient-Based Learning Applied to Document Recognition","authors":"Yann Lecun, Léon Bottou, Yoshua Bengio, Patrick Haffner","year":"1998","venue":"Proc. of the IEEE","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Lindsay et al., 2017] Alan Lindsay, Jonathon Read, Joao F Ferreira, et al. Framer: Planning Models from Natural Lan- guage Action Descriptions. In ICAPS, 2017.","title":"Framer: Planning Models from Natural Language Action Descriptions","authors":"Alan Lindsay, Jonathon Read, Joao F Ferreira","year":"2017","venue":"ICAPS","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Liu et al., 2019] Liyuan Liu, Haoming Jiang, Pengcheng He, et al. On the Variance of the Adaptive Learning Rate and Beyond. arXiv:1908.03265, 2019.","title":"On the Variance of the Adaptive Learning Rate and Beyond","authors":"Liyuan Liu, Haoming Jiang, Pengcheng He","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Maddison et al., 2017] Chris J. Maddison, Andriy Mnih, and Yee Whye Teh. The Concrete Distribution: A Contin- uous Relaxation of Discrete Random Variables. In ICLR, 2017.","title":"The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables","authors":"Chris J Maddison, Andriy Mnih, Yee Whye Teh","year":"2017","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Mnih et al., 2015] Volodymyr Mnih, Koray Kavukcuoglu, David Silver, et al. Human-Level Control through Deep Reinforcement Learning. Nature, 518(7540), 2015.","title":"Human-Level Control through Deep Reinforcement Learning","authors":"Volodymyr Mnih, Koray Kavukcuoglu, David Silver","year":"2015","venue":"Nature","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Payan, 1992] Charles Payan. On the Chromatic Number of Cube-Like Graphs. Discrete mathematics, 103(3), 1992.","title":"On the Chromatic Number of Cube-Like Graphs","authors":"Charles Payan","year":"1992","venue":"Discrete mathematics","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Richter and Westphal, 2010] Silvia Richter and Matthias Westphal. The LAMA Planner: Guiding Cost-Based Any- time Planning with Landmarks. J. Artif. Intell. Res.(JAIR), 39(1), 2010.","title":"The LAMA Planner: Guiding Cost-Based Anytime Planning with Landmarks","authors":"Silvia Richter, Matthias Westphal","year":"2010","venue":"J. Artif. Intell. Res.(JAIR)","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Silver et al., 2016] David Silver, Aja Huang, Chris J Mad- dison, et al. Mastering the Game of Go with Deep Neural Networks and Tree Search. nature, 529(7587):484, 2016.","title":"Mastering the Game of Go with Deep Neural Networks and Tree Search","authors":"David Silver, Aja Huang, Chris J Maddison","year":"2016","venue":"nature","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Taddeo and Floridi, 2005] Mariarosaria Taddeo and Lu- ciano Floridi. Solving the Symbol Grounding Problem: A Critical Review of Fifteen Years of Research. Journal of Experimental & Theoretical Artificial Intelligence, 17(4), 2005.","title":"Mariarosaria Taddeo and Luciano Floridi. Solving the Symbol Grounding Problem: A Critical Review of Fifteen Years of Research","authors":"Taddeo, Floridi","year":"2005","venue":"Journal of Experimental & Theoretical Artificial Intelligence","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Vizing, 1965] Vadim G Vizing. The Chromatic Class of a Multigraph. Cybernetics and Systems Analysis, 1(3), 1965.","title":"The Chromatic Class of a Multigraph","authors":"G Vadim, Vizing","year":"1965","venue":"Cybernetics and Systems Analysis","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[Yang et al., 2007] Qiang Yang, Kangheng Wu, and Yunfei Jiang. Learning Action Models from Plan Examples us- ing Weighted MAX-SAT. Artificial Intelligence, 171(2-3), 2007.","title":"Learning Action Models from Plan Examples using Weighted MAX-SAT","authors":"Qiang Yang, Kangheng Wu, Yunfei Jiang","year":"2007","venue":"Artificial Intelligence","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}