Argumentative Machine Learning

Project description

Classification is the problem of categorizing new observations by using a classifier learnt from already categorized examples. In general, the area of machine learning has brought forth a series of different approaches to deal with this problem, from decision trees to support vector machines and others. Recently, approaches to statistical relational learning even take the perspective of knowledge representation and reasoning into account by developing models on more formal logical and statistical grounds. In this project, we will significantly generalize this reasoning aspect of machine learning towards the use of computational models of argumentation, a popular approach to commonsense reasoning, for reasoning within machine learning. Consider e.g. the following two-step classification approach. In the first step, rule learning algorithms are used to extract frequent patterns and rules from a given data set. The output of this step comprises a huge number of rules (given fairly low confidence and support parameters) and these cannot directly be used for the purpose of classification as they are usually inconsistent with one another. Therefore, in the second step, we interpret these rules as the input for approaches to structured argumentation - more specifically ASPIC+, DeLP, ABA, and deductive argumentation - and probabilistic and other quantitative extensions of those. Using the argumentative inference procedures of these approaches and given a new observation, the classification of the new observation is determined by constructing arguments on top of these rules for the different classes and determining their justification status. More precisely, the project CAML will investigate radically novel machine learning approaches as the one outlined above in detail and develop the new field of "Argumentative Machine Learning" in general: a tight integration of "C"omputational "A"rgumentation und "M"achine "L"earning. This has several benefits. The use of argumentation techniques allows to obtain classifiers, which are by design able to explain their decisions, and therefore addresses the recent need for Explainable AI: classifications are accompanied by a dialectical analysis showing why arguments for the conclusion are preferred to counterarguments; this automatic deliberation, validation, reconstruction and synthesis of arguments helps in assessing trust in the classifier, which is fundamental if one plans to take action based on a prediction. Argumentation techniques in machine learning also allows the easy integration of additional expert knowledge in form of arguments. As there are many different approaches to structured argumentation that take different perspectives on the issue of argumentation, their application in machine learning will provide new insights on their usefulness and allows for a comparison between them on a different level.


Project leader

People

Project duration

October 2018 - July 2021

Follow-up project

CAML2 - Causality, Argumentation, and Machine Learning

Publications

  • Anthony Hunter, Sylwia Polberg, Nico Potyka, Tjitze Rienstra, Matthias Thimm. Probabilistic Argumentation: A Survey. In Dov Gabbay, Massimiliano Giacomin, Guillermo R. Simari, Matthias Thimm (Eds.), Handbook of Formal Argumentation, College Publications. August 2021. bibtex pdf
  • Isabelle Kuhlmann, Tjitze Rienstra, Lars Bengel, Kenneth Skiba, Matthias Thimm. Distinguishability in Abstract Argumentation. In Proceedings of the 18th International Conference on Principles of Knowledge Representation and Reasoning (KR'21). November 2021. bibtex pdf
  • Kenneth Skiba, Tjitze Rienstra, Matthias Thimm, Jesse Heyninck, Gabriele Kern-Isberner. Ranking Extensions in Abstract Argumentation. In Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI'21). August 2021. bibtex pdf
  • Matthias Thimm, Federico Cerutti, Mauro Vallati. Skeptical Reasoning with Preferred Semantics in Abstract Argumentation without Computing Preferred Extensions. In Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI'21). August 2021. bibtex pdf
  • Matthias Thimm, Federico Cerutti, Mauro Vallati. Fudge: A Light-weight Solver for Abstract Argumentation based on SAT Reductions. In The Fourth International Competition on Computational Models of Argumentation (ICCMA'21). May 2021. bibtex pdf
  • Matthias Thimm. Harper++: Using Grounded Semantics for Approximate Reasoning in Abstract Argumentation. In The Fourth International Competition on Computational Models of Argumentation (ICCMA'21). May 2021. bibtex pdf
  • Xiaoting Shao, Arseny Skryagin, Patrick Schramowski, Wolfgang Stammer, Kristian Kersting (2021): Right for Better Reasons: Training Differentiable Models by Constraining their Influence Function. In Proceedings of Thirty-Fifth AAAI Conference on Artificial Intelligence (AAAI). pdf
  • Tjitze Rienstra, Matthias Thimm, Xiaoting Shao, Kristian Kersting. Independence and D-separation in Abstract Argumentation. In Proceedings of the 17th International Conference on Principles of Knowledge Representation and Reasoning (KR). September 2020. bibtex pdf
  • Xiaoting Shao, Zhongjie Yu, Arseny Skryagin, Tjitze Rienstra, Matthias Thimm, Kristian Kersting. Modelling Multivariate Ranking Functions with Min-Sum Networks. In Proceedings of the 14th International Conference on Scalable Uncertainty Management (SUM'20). September 2020. bibtex pdf
  • Matthias Thimm, Federico Cerutti, Mauro Vallati. On Computing the Set of Acceptable Arguments in Abstract Argumentation. In Proceedings of the 8th International Conference on Computational Models of Argument (COMMA'20). September 2020. bibtex pdf
  • Jonas Klein, Matthias Thimm. Revisiting SAT Techniques for Abstract Argumentation. In Proceedings of the 8th International Conference on Computational Models of Argument (COMMA'20). September 2020. bibtex pdf
  • Matthias Thimm, Tjitze Rienstra. Approximate Reasoning with ASPIC+ by Argument Sampling. In Proceedings of the Third International Workshop on Systems and Algorithms for Formal Argumentation (SAFA'20). September 2020. bibtex pdf
  • Xiaoting Shao, Tjitze Rienstra, Matthias Thimm, Kristian Kersting. Towards Understanding and Arguing with Classifiers: Recent Progress. In Datenbank-Spektrum. June 2020. bibtex pdf
  • Andrea Galassi, Kristian Kersting, Marco Lippi, Xiaoting Shao, Paolo Torroni. Neural-Symbolic Argumentation Mining: An Argument in Favor of Deep Learning and Reasoning. In Frontiers in Big Data, https://doi.org/10.3389/fdata.2019.00052.s001, 2020.
  • Patrick Schramowski, Wolfgang Stammer, Stefano Teso, Anna Brugger, Franziska Herbet, Xiaoting Shao, Hans-Georg Luigs, Anne-Katrin Mahlein, Kristian Kersting. Right for the Wrong Scientific Reasons: Revising Deep Networks by Interacting with their Explanations. arXiv preprint arXiv:2001.05371, 2020.
  • Anthony Hunter, Sylwia Polberg, Matthias Thimm. Epistemic Graphs for Representing and Reasoning with Positive and Negative Influences of Arguments. In Artificial intelligence. January 2020. bibtex pdf
  • Isabelle Kuhlmann, Matthias Thimm. Using Graph Convolutional Networks for Approximate Reasoning with Abstract Argumentation Frameworks: A Feasibility Study. In Proceedings of the 13th International Conference on Scalable Uncertainty Management (SUM'19). December 2019. bibtex pdf
  • Tjitze Rienstra. Ranked Programming. In Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI'19), August 2019. bibtex pdf
  • Robert Peharz, Antonio Vergari, Karl Stelzner, Alejandro Molina, Xiaoting Shao, Martin Trapp, Kristian Kersting, Zoubin Ghahramani. Random Sum-Product Networks: A Simple and Effective Approach to Probabilistic Deep Learning. In Proceedings of the 35th Conference on Uncertainty in Artificial Intelligence (UAI), 2019.
  • Federico Cerutti, Matthias Thimm. A General Approach to Reasoning with Probabilities. In International Journal of Approximate Reasoning, 111:35-50. August 2019. bibtex pdf
  • Antonio Vergari, Alejandro Molina, Robert Peharz, Zoubin Ghahramani, Kristian Kersting, Isabel Valera. Automatic Bayesian Density Analysis. In proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI'19), 2019.
  • Stefano Teso, Kristian Kersting. Explanatory Interactive Machine Learning. In Proceedings of the 2nd AAAI/ACM Conference on AI, Ethics, and Society (AIES). 2019
  • Federico Cerutti, Matthias Thimm. Probabilistic Augmentations for Knowledge Representation Formalisms. In Proceedings of the 2018 Workshop on Hybrid Reasoning and Learning (HRL'18). October 2018. bibtex pdf
  • Tjitze Rienstra, Matthias Thimm, Beishui Liao, Leendert van der Torre. Probabilistic Abstract Argumentation based on SCC Decomposability. In Proceedings of the 16th International Conference on Principles of Knowledge Representation and Reasoning (KR'19). October 2018. bibtex pdf
  • Federico Cerutti, Matthias Thimm. A General Approach to Reasoning with Probabilities (Extended Abstract). In Proceedings of the 16th International Conference on Principles of Knowledge Representation and Reasoning (KR'18). October 2018. bibtex pdf
  • Matthias Thimm, Sylwia Polberg, Anthony Hunter. Epistemic Attack Semantics. In Proceedings of the Seventh International Conference on Computational Models of Argumentation (COMMA'18). September 2018. bibtex pdf
  • Matthias Thimm, Federico Cerutti, Tjitze Rienstra. Probabilistic Graded Semantics. In Proceedings of the Seventh International Conference on Computational Models of Argumentation (COMMA'18). September 2018. bibtex pdf
  • Tjitze Rienstra, Matthias Thimm. Ranking Functions over Labellings. In Proceedings of the Seventh International Conference on Computational Models of Argumentation (COMMA'18). September 2018. bibtex pdf
  • Matthias Thimm. Stochastic Local Search Algorithms for Abstract Argumentation under Stable Semantics. In Proceedings of the Seventh International Conference on Computational Models of Argumentation (COMMA'18). September 2018. bibtex pdf
  • Leendert van der Torre, Tjitze Rienstra, Dov Gabbay. Argumentation as Exogenous Coordination. In Frank de Boer, Marcello Bonsangue, and Jan Rutten, editors, It's All About Coordination: Essays to Celebrate the Lifelong Scientific Achievements of Farhad Arbab. Springer International Publishing, 2018.
  • Matthias Thimm, Kristian Kersting. Towards Argumentation-based Classification. In Logical Foundations of Uncertainty and Machine Learning, Workshop at IJCAI'17. August 2017. bibtex pdf



Last updated 18.05.2021, Matthias Thimm | Terms