{"type":"ScholarlyArticle","title":"Towards a Theory of Machine Learning on Graphs and its Applications in Combinatorial Optimization","authors":[{"name":"Christopher Morris","slug":"christopher-morris","given":"Christopher","family":"Morris","affiliation":"RWTH Aachen University"}],"abstract":"Machine learning on graphs, especially using graph neural networks (GNNs), has seen a surge in interest due to the wide availability of graph data across many disciplines, from life and physical to social and engineering sciences. Despite their practical success, our theoretical understanding of the properties of GNNs remains incomplete. Here, we survey the author's and his collaborators' progress in developing a deeper theoretical understanding of GNNs' expressive power and generalization abilities. In addition, we overview recent progress in using GNNs to speed up solvers for hard combinatorial optimization tasks.","doi":"10.24963/ijcai.2024/981","url":"https://ijcai.org/proceedings/2024/981","pdf":"https://ijcai.org/proceedings/2024/0981.pdf","markdown":"https://ijcai.org/proceedings/2024/981.md","bibtex":"https://ijcai.org/proceedings/2024/bibtex/981","volume":{"slug":"2024","title":"Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2024","url":"https://ijcai.org/proceedings/2024/","year":2024,"location":"Jeju, Korea","dates":"3-9 August 2024","editors":"Kate Larson","isbn":"978-1-956792-04-1","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2024"},"section":"Early Career","track":"Early Career","number":981,"pages":{"first":8553,"last":8558},"keywords":["Machine Learning: ML: Representation learning","Machine Learning: ML: Theory of deep learning","Constraint Satisfaction and Optimization: CSO: Constraint optimization problems","Machine Learning: ML: Learning theory"],"videos":[],"references":[{"position":1,"raw":"A. Aamand, J. Y. Chen, P. Indyk, S. Narayanan, R. Rubinfeld, N. Schiefer, S. Silwal, and T. Wagner. Exponentially improv- ing the complexity of simulating the Weisfeiler-Lehman test with graph neural networks. arXiv preprint, 2022.","title":"Exponentially improving the complexity of simulating the Weisfeiler-Lehman test with graph neural networks","authors":"A Aamand, J Y Chen, P Indyk, S Narayanan, R Rubinfeld, N Schiefer, S Silwal, T Wagner","year":"2022","venue":null,"doi":null,"url":null,"internal_url":null},{"position":2,"raw":"T. Amir, S. J. Gortler, I. Avni, R. Ravina, and N. Dym. Neural injective functions for multisets, measures and graphs via a finite witness theorem. arXiv preprint, 2023.","title":"Neural injective functions for multisets, measures and graphs via a finite witness theorem","authors":"T Amir, S J Gortler, I Avni, R Ravina, N Dym","year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":3,"raw":"V. Arvind, J. Köbler, G. Rattan, and O. Verbitsky. On the power of color refinement. In International Symposium on Fundamentals of Computation Theory, pages 339-350, 2015.","title":"On the power of color refinement","authors":"V Arvind, J Köbler, G Rattan, O Verbitsky","year":"2015","venue":"International Symposium on Fundamentals of Computation Theory","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"W. Azizian and M. Lelarge. Characterizing the expressive power of invariant and equivariant graph neural networks. In ICLR, 2021.","title":"Characterizing the expressive power of invariant and equivariant graph neural networks","authors":"W Azizian, M Lelarge","year":"2021","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"M. Bechler-Speicher, I. Amos, R. Gilad-Bachrach, and A. Globerson. Graph neural networks use graphs when they shouldn't. arXiv preprint, 2023.","title":"Graph neural networks use graphs when they shouldn't","authors":"M Bechler-Speicher, I Amos, R Gilad-Bachrach, A Globerson","year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"L. Bergen, T. J. O'Donnell, and D. Bahdanau. Systematic generalization with edge transformers. In NeurIPS, 2021.","title":"Systematic generalization with edge transformers","authors":"L Bergen, T J O'donnell, D Bahdanau","year":"2021","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"J. Böker. Graph similarity and homomorphism densities. In ICALP, 2021.","title":"Graph similarity and homomorphism densities","authors":"J Böker","year":"2021","venue":"ICALP","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"G. Bouritsas, F. Frasca, S. Zafeiriou, and M. M. Bronstein. Improving graph neural network expressivity via subgraph isomorphism counting. arXiv preprint, 2020.","title":"Improving graph neural network expressivity via subgraph isomorphism counting","authors":"G Bouritsas, F Frasca, S Zafeiriou, M M Bronstein","year":"2020","venue":null,"doi":null,"url":null,"internal_url":null},{"position":9,"raw":"C. Bravo, Al. Kozachinskiy, and C. Rojas. On dimensionality of feature vectors in mpnns. arXiv preprint, 2024.","title":"On dimensionality of feature vectors in mpnns","authors":"C Bravo, Al Kozachinskiy, C Rojas","year":"2024","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"J. Böker, R. Levie, N. Huang, S. Villar, and C. Morris. Fine- grained expressivity of graph neural networks. In NeurIPS, 2023.","title":"Finegrained expressivity of graph neural networks","authors":"J Böker, R Levie, N Huang, S Villar, C Morris","year":"2023","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"J. Cai, M. Fürer, and N. Immerman. An optimal lower bound on the number of variables for graph identifications. Combi- natorica, 12(4):389-410, 1992.","title":"An optimal lower bound on the number of variables for graph identifications","authors":"J Cai, M Fürer, N Immerman","year":"1992","venue":"Combinatorica","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"Q. Cappart, D. Chételat, E. B. Khalil, A. Lodi, C. Morris, and P. Veličković. Combinatorial optimization and reasoning with graph neural networks. In IJCAI, 2021.","title":"Combinatorial optimization and reasoning with graph neural networks","authors":"Q Cappart, D Chételat, E B Khalil, A Lodi, C Morris, P Veličković","year":"2021","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2021/595"},{"position":13,"raw":"S. Chen, S. Lim, F. Mémoli, Z. Wan, and Y. Wang. Weisfeiler- Lehman meets Gromov-Wasserstein. In ICML, 2022.","title":"Weisfeiler-Lehman meets Gromov-Wasserstein","authors":"S Chen, S Lim, F Mémoli, Z Wan, Y Wang","year":"2022","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"M. Colombi, R. Mansini, and M. Savelsbergh. The generalized independent set problem: Polyhedral analysis and solution approaches. European Journal of Operational Research, 260(1):41-55, 2017.","title":"The generalized independent set problem: Polyhedral analysis and solution approaches","authors":"M Colombi, R Mansini, M Savelsbergh","year":"2017","venue":"European Journal of Operational Research","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"S. S. Du, K. Hou, R. Salakhutdinov, B. Póczos, Ruosong Wang, and K. Xu. Graph neural tangent kernel: Fusing graph neural networks with graph kernels. In NeurIPS, 2019.","title":"Graph neural tangent kernel: Fusing graph neural networks with graph kernels","authors":"S S Du, K Hou, R Salakhutdinov, B Póczos, Ruosong Wang, K Xu","year":"2019","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"N. Dziri, X. Lu, M. Sclar, X. Lorraine Li, L. Jiang, B. Yuchen Lin, S. Welleck, P. West, C. Bhagavatula, R. Le Bras, J. D. Hwang, S. Sanyal, X. Ren, A. Ettinger, Z. Harchaoui, and Y. Choi. Faith and fate: Limits of transformers on composi- tionality. In NeurIPS, 2023.","title":"Faith and fate: Limits of transformers on compositionality","authors":"N Dziri, X Lu, M Sclar, X Li, L Jiang, B Yuchen Lin, S Welleck, P West, C Bhagavatula, R Le Bras, J D Hwang, S Sanyal, X Ren, A Ettinger, Z Harchaoui, Y Choi","year":"2023","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"D. Easley and J. Kleinberg. Networks, Crowds, and Markets: Reasoning About a Highly Connected World. Cambridge University Press, 2010.","title":"Networks, Crowds, and Markets: Reasoning About a Highly Connected World","authors":"D Easley, J Kleinberg","year":"2010","venue":null,"doi":null,"url":null,"internal_url":null},{"position":18,"raw":"B. J. Franks, C. Morris, A. Velingker, and F. Geerts. Weisfeiler- leman at the margin: When more expressivity matters. In ICML, 2024.","title":"Weisfeilerleman at the margin: When more expressivity matters","authors":"B J Franks, C Morris, A Velingker, F Geerts","year":"2024","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"V. K. Garg, S. Jegelka, and T. S. Jaakkola. Generalization and representational limits of graph neural networks. In ICLR, 2020.","title":"Generalization and representational limits of graph neural networks","authors":"V K Garg, S Jegelka, T S Jaakkola","year":"2020","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"M. Gasse, D. Chételat, N. Ferroni, L. Charlin, and A. Lodi. Exact combinatorial optimization with graph convolutional neural networks. In NeurIPS, 2019.","title":"Exact combinatorial optimization with graph convolutional neural networks","authors":"M Gasse, D Chételat, N Ferroni, L Charlin, A Lodi","year":"2019","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"F. Geerts and J. L. Reutter. Expressiveness and approximation properties of graph neural networks. In ICLR, 2022.","title":"Expressiveness and approximation properties of graph neural networks","authors":"F Geerts, J L Reutter","year":"2022","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"J. Gilmer, S. S. Schoenholz, P. F. Riley, O. Vinyals, and G. E. Dahl. Neural message passing for quantum chemistry. In ICLR, 2017.","title":"Neural message passing for quantum chemistry","authors":"J Gilmer, S S Schoenholz, P F Riley, O Vinyals, G E Dahl","year":"2017","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"J. Gondzio. Interior point methods 25 years later. European Journal of Operational Research, 218:587-601, 2012.","title":"Interior point methods 25 years later","authors":"J Gondzio","year":"2012","venue":"European Journal of Operational Research","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"M. Grohe. The logic of graph neural networks. In Symposium on Logic in Computer Science, pages 1-17, 2021.","title":"The logic of graph neural networks","authors":"M Grohe","year":"2021","venue":"Symposium on Logic in Computer Science","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"M. Hewitt, G. L. Nemhauser, and M. W. P. Savelsbergh. Com- bining exact and heuristic approaches for the capacitated fixed-charge network flow problem. INFORMS Journal on Computing, 22(2):314-325, 2010.","title":"Combining exact and heuristic approaches for the capacitated fixed-charge network flow problem","authors":"M Hewitt, G L Nemhauser, M W P Savelsbergh","year":"2010","venue":"INFORMS Journal on Computing","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"E. I. Hsu, C. J. Muise, J. C. Beck, and S. A. McIlraith. Proba- bilistically estimating backbones and variable bias: Experi- mental overview. In International Conference on Principles and Practice of Constraint Programming, 2008.","title":"Probabilistically estimating backbones and variable bias: Experimental overview","authors":"E I Hsu, C J Muise, J C Beck, S A Mcilraith","year":"2008","venue":"International Conference on Principles and Practice of Constraint Programming","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"IBM. CPLEX User's Manual, Version 12.10.0, 2020.","title":"CPLEX User's Manual","authors":null,"year":"2020","venue":"Version","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"E. B. Khalil, C. Morris, and A. Lodi. MIP-GNN: A data-driven framework for guiding combinatorial solvers. In AAAI, pages 10219-10227, 2022.","title":"MIP-GNN: A data-driven framework for guiding combinatorial solvers","authors":"E B Khalil, C Morris, A Lodi","year":"2022","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"J. Kim, S. Oh, and S. Hong. Transformers generalize deepsets and can be extended to graphs & hypergraphs. In NeurIPS, 2021.","title":"Transformers generalize deepsets and can be extended to graphs & hypergraphs","authors":"J Kim, S Oh, S Hong","year":"2021","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"J. Kim, D. Nguyen, S. Min, S. Cho, M. Lee, H. Lee, and S. Hong. Pure transformers are powerful graph learners. In NeurIPS, 2022.","title":"Pure transformers are powerful graph learners","authors":"J Kim, D Nguyen, S Min, S Cho, M Lee, H Lee, S Hong","year":"2022","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":31,"raw":"D. P. Kingma and J. Ba. Adam: A method for stochastic optimization. In ICLR, 2015.","title":"Adam: A method for stochastic optimization","authors":"D P Kingma, J Ba","year":"2015","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":32,"raw":"M. Mohri, A. Rostamizadeh, and A. Talwalkar. Foundations of Machine Learning. MIT Press, 2012.","title":"Foundations of Machine Learning","authors":"M Mohri, A Rostamizadeh, A Talwalkar","year":"2012","venue":null,"doi":null,"url":null,"internal_url":null},{"position":33,"raw":"C. Morris, M. Ritzert, M. Fey, W. L. Hamilton, J. E. Lenssen, G. Rattan, and M. Grohe. Weisfeiler and Leman go neural: Higher-order graph neural networks. In AAAI, 2019.","title":"Weisfeiler and Leman go neural: Higher-order graph neural networks","authors":"C Morris, M Ritzert, M Fey, W L Hamilton, J E Lenssen, G Rattan, M Grohe","year":"2019","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":34,"raw":"C. Morris, G. Rattan, and P. Mutzel. Weisfeiler and Leman go sparse: Towards higher-order graph embeddings. In NeurIPS, 2020.","title":"Weisfeiler and Leman go sparse: Towards higher-order graph embeddings","authors":"C Morris, G Rattan, P Mutzel","year":"2020","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":35,"raw":"C. Morris, F. Geerts, J. Tönshoff, and M. Grohe. WL meet VC. In ICML, 2023.","title":"ICML","authors":"C Morris, F Geerts, J Tönshoff, M Grohe, Wl, Vc","year":"2023","venue":null,"doi":null,"url":null,"internal_url":null},{"position":36,"raw":"C. Morris, Y. L., H. Maron, B. Rieck, N. M. Kriege, M. Grohe, M. Fey, and K. Borgwardt. Weisfeiler and Leman go ma- chine learning: The story so far. JMLR, 2023.","title":"Weisfeiler and Leman go machine learning: The story so far","authors":"C Morris, Y L, H Maron, B Rieck, N M Kriege, M Grohe, M Fey, K Borgwardt","year":"2023","venue":"JMLR","doi":null,"url":null,"internal_url":null},{"position":37,"raw":"C. Morris, F. Frasca, N. Dym, H. Maron, İ. İ. Ceylan, R. Levie, D. Lim, M. M. Bronstein, M. Grohe, and S. Jegelka. Future directions in the theory of graph machine learning. In ICML, 2024.","title":"Future directions in the theory of graph machine learning","authors":"C Morris, F Frasca, N Dym, H Maron, İ İ Ceylan, R Levie, D Lim, M M Bronstein, M Grohe, S Jegelka","year":"2024","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":38,"raw":"L. Müller, M. Galkin, C. Morris, and L. Rampásek. Attending to graph transformers. TMLR, 2024.","title":"Attending to graph transformers","authors":"L Müller, M Galkin, C Morris, L Rampásek","year":"2024","venue":"TMLR","doi":null,"url":null,"internal_url":null},{"position":39,"raw":"L. Müller and C. Morris. Aligning transformers with weisfeiler- leman. In ICML, 2024.","title":"Aligning transformers with weisfeilerleman","authors":"L Müller, C Morris","year":"2024","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":40,"raw":"J. Nocedal and S. J. Wright. Numerical optimization. Springer, 2 edition, 2006.","title":"Numerical optimization","authors":"J Nocedal, S J Wright","year":"2006","venue":null,"doi":null,"url":null,"internal_url":null},{"position":41,"raw":"C. Qian, G. Rattan, F. Geerts, C. Morris, and M. Niepert. Or- dered subgraph aggregation networks. In NeurIPS, 2022.","title":"Ordered subgraph aggregation networks","authors":"C Qian, G Rattan, F Geerts, C Morris, M Niepert","year":"2022","venue":"NeurIPS","doi":null,"url":null,"internal_url":null},{"position":42,"raw":"C. Qian, D. Chételat, and C. Morris. Exploring the power of graph neural networks in solving linear optimization prob- lems. In AISTATS, 2024.","title":"Exploring the power of graph neural networks in solving linear optimization problems","authors":"C Qian, D Chételat, C Morris","year":"2024","venue":"AISTATS","doi":null,"url":null,"internal_url":null},{"position":43,"raw":"F. Scarselli, M. Gori, A. C. Tsoi, M. Hagenbuchner, and G. Monfardini. The graph neural network model. IEEE Transactions on Neural Networks, 20(1):61-80, 2009.","title":"The graph neural network model","authors":"F Scarselli, M Gori, A C Tsoi, M Hagenbuchner, G Monfardini","year":"2009","venue":"IEEE Transactions on Neural Networks","doi":null,"url":null,"internal_url":null},{"position":44,"raw":"F. Scarselli, A. C. Tsoi, and M. Hagenbuchner. The Vapnik- Chervonenkis dimension of graph and recursive neural net- works. Neural Networks, pages 248-259, 2018.","title":"The Vapnik-Chervonenkis dimension of graph and recursive neural networks","authors":"F Scarselli, A C Tsoi, M Hagenbuchner","year":"2018","venue":"Neural Networks","doi":null,"url":null,"internal_url":null},{"position":45,"raw":"D. F. Shanno. Who invented the interior-point method? (https://doi.org/10.4171/dms/6/11), 2012.","title":"Who invented the interior-point method?","authors":"D F Shanno","year":null,"venue":null,"doi":"10.4171/dms/6/11),2012","url":"https://doi.org/10.4171/dms/6/11),2012","internal_url":null},{"position":46,"raw":"J. Stokes, K. Yang, K. Swanson, W. Jin, A. Cubillos-Ruiz, N. Donghia, C. MacNair, S. French, L. Carfrae, Z. Bloom- Ackerman, V. Tran, A. Chiappino-Pepe, A. Badran, I. An- drews, E. Chory, G. Church, E. Brown, T. Jaakkola, R. Barzi- lay, and J. Collins. A deep learning approach to antibiotic discovery. Cell, pages 688-702.e13, 2020.","title":"A deep learning approach to antibiotic discovery","authors":"J Stokes, K Yang, K Swanson, W Jin, A Cubillos-Ruiz, N Donghia, C Macnair, S French, L Carfrae, Z Bloom-Ackerman, V Tran, A Chiappino-Pepe, A Badran, I Andrews, E Chory, G Church, E Brown, T Jaakkola, R Barzilay, J Collins","year":"2020","venue":"Cell","doi":null,"url":null,"internal_url":null},{"position":47,"raw":"H. Tang and Y. Liu. Towards understanding generalization of graph neural networks. In ICML, 2023.","title":"Towards understanding generalization of graph neural networks","authors":"H Tang, Y Liu","year":"2023","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":48,"raw":"V. N. Vapnik and A. Chervonenkis. A note on one class of perceptrons. Avtomatika i Telemekhanika, 24(6):937-945, 1964.","title":"A note on one class of perceptrons","authors":"V N Vapnik, A Chervonenkis","year":"1964","venue":"Avtomatika i Telemekhanika","doi":null,"url":null,"internal_url":null},{"position":49,"raw":"V. N. Vapnik. The Nature of Statistical Learning Theory. Springer, 1995.","title":"The Nature of Statistical Learning Theory","authors":"V N Vapnik","year":"1995","venue":null,"doi":null,"url":null,"internal_url":null},{"position":50,"raw":"V. Vapnik. Statistical learning theory. Wiley, 1998.","title":"Statistical learning theory","authors":"V Vapnik","year":"1998","venue":null,"doi":null,"url":null,"internal_url":null},{"position":51,"raw":"P. Velickovic, A. P. Badia, D. Budden, R. Pascanu, A. Banino, Dashevskiy M, Hadsell R, and C. Blundell. The CLRS algorithmic reasoning benchmark. In ICML, 2022.","title":"The CLRS algorithmic reasoning benchmark","authors":"P Velickovic, A P Badia, D Budden, R Pascanu, A Banino, M Dashevskiy, R Hadsell, C Blundell","year":"2022","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":52,"raw":"B. Weisfeiler and A. Leman. The reduction of a graph to canon- ical form and the algebra which appears therein. Nauchno- Technicheskaya Informatsia, 2(9):12-16, 1968.","title":"The reduction of a graph to canonical form and the algebra which appears therein","authors":"B Weisfeiler, A Leman","year":"1968","venue":"Nauchno-Technicheskaya Informatsia","doi":null,"url":null,"internal_url":null},{"position":53,"raw":"F. Wong, E. J. Zheng, J. A. Valeri, N. M. Donghia, Melis N. Anahtar, Satotaka Omori, Alicia Li, Andres Cubillos-Ruiz, Aarti Krishnan, Wengong Jin, Abigail L. Manson, Jens Friedrichs, Ralf Helbig, Behnoush Hajian, Dawid K. Fiejtek, Florence F. Wagner, Holly H. Soutter, Ashlee M. Earl, Jonathan M. Stokes, Lars D. Renner, and James J. Collins. Discovery of a structural class of antibiotics with explainable deep learning. Nature, 2023.","title":"Discovery of a structural class of antibiotics with explainable deep learning","authors":"F Wong, E J Zheng, J A Valeri, N M Donghia, N Melis, Satotaka Anahtar, Alicia Omori, Andres Li, Aarti Cubillos-Ruiz, Wengong Krishnan, Abigail L Jin, Jens Manson, Ralf Friedrichs, Behnoush Helbig, Hajian, K Dawid, Florence F Fiejtek, Holly H Wagner, Ashlee M Soutter, Jonathan M Earl, Lars D Stokes, James J Renner, Collins","year":"2023","venue":"Nature","doi":null,"url":null,"internal_url":null},{"position":54,"raw":"K. Xu, W. Hu, J. Leskovec, and S. Jegelka. How powerful are graph neural networks? In ICLR, 2019.","title":"How powerful are graph neural networks?","authors":"K Xu, W Hu, J Leskovec, S Jegelka","year":"2019","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":55,"raw":"K. Xu, M. Zhang, S. Jegelka, and K. Kawaguchi. Optimiza- tion of graph neural networks: Implicit acceleration by skip connections and more depth. In ICLR, 2021.","title":"Optimization of graph neural networks: Implicit acceleration by skip connections and more depth","authors":"K Xu, M Zhang, S Jegelka, K Kawaguchi","year":"2021","venue":"ICLR","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}