Abstract
Statistical artificial intelligence based upon machine learning is facing major challenges such as machine bias, explainability, and verifiability problems. Resolving them would be of the utmost importance for ethical, safe, and responsible AI for social good. In this paper we propose to address these problems with the (recently rapidly developing) methodology of category theory, a powerful integrative scientific language from pure mathematics, and discuss, in particular, the possibility of the categorical integration of statistical (inductive) with symbolic (deductive) artificial intelligence. Categorical artificial intelligence arguably has the potential to resolve the aforementioned urgent problems, thus being beneficial to make artificial intelligence more ethical, verifiable, responsible, and so more human, which would be crucial for AI-pervasive society.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Arrieta, A.B., et al.: Explainable artificial intelligence (XAI). Inf. Fus. 58, 82–115 (2020)
Baez, J.: Physics, topology, logic and computation: a Rosetta stone. Lect. Notes Phys. 813, 95–174 (2011)
Baez, J., Pollard, B.: A compositional framework for reaction networks. Rev. Math. Phys. 29, 1750028 (2017)
Besold, T.R., et al.: Neural-Symbolic Learning and Reasoning: A Survey and Interpretation. arXiv:1711.03902 (2017)
Caliskan, A., et al.: Semantics derived automatically from language corpora contain human-like biases. Science 356, 183–186 (2017)
Chollet, F.: On the Measure of Intelligence. arXiv:1911.01547 (2019)
Coecke, B.: Automated quantum reasoning: non-logic - semi-logic - hyper-logic. In: Proceedings of Quantum Interaction, pp. 31–38 (2007)
Coecke, B.: Quantum picturalism. Contemp. Phys. 51, 59–83 (2009)
Coecke, B., et al.: Mathematical foundations for a compositional distributional model of meaning. Linguist. Anal. 36, 345–384 (2010)
Coecke, B.: The logic of quantum mechanics - Take II. arXiv:1204.3458 (2012)
Coecke, B., Kissinger, A.: Picturing quantum processes. In: CUP (2017)
Cruttwell, G., et al.: Categorical Foundations of Gradient-Based Learning. arXiv:2103.01931 (2021)
de Felice, G., et al.: Discopy: monoidal categories in Python. In: Proceedings of ACT 2020, pp. 183–197 (2020)
Domingos, P., Lowd, D.: Unifying logical and statistical AI with Markov logic. Commun. ACM 62, 74–83 (2019)
Fong, B., et al.: Backprop as functor. In: Proceedings of LICS 2019, pp. 1–13 (2019)
Fong, B., Spivak, D.I.: An invitation to applied category theory: seven sketches in compositionality. In: CUP (2019)
Fritz, T.: A synthetic approach to Markov kernels, conditional independence and theorems on sufficient statistics. Adv. Math. 370, 107239 (2020)
Gaboardi, M., et al.: Graded Hoare Logic and its Categorical Semantics. arXiv:2007.11235 (2020)
Grefenstette, E., Sadrzadeh, M.: Experimental support for a categorical compositional distributional model of meaning. In: Proceedings of EMNLP’11, pp. 1394–1404 (2011)
Jacobs, B.: Categorical Logic and Type Theory. Elsevier, Amsterdam (1999)
Kartsaklis, D.: Tensor-based models of natural language semantics. In: Lecture Slides at the Workshop on Tensors, their Decomposition, and Applications (2016)
Kashiawara, M., Schapira, P.: Categories and Sheaves. Springer, Heidelberg (2006). https://doi.org/10.1007/3-540-27950-4
Lawvere, F.W.: Adjointness in foundations. Dialectica 23, 281–296 (1969)
Lorenz, R., et al.: QNLP in practice. arXiv:2102.12846 (2021)
Maruyama, Y.: Dualities for algebras of Fitting’s many-valued modal logics. Fundamenta Informaticae 106, 273–294 (2011)
Maruyama, Y.: Natural duality, modality, and coalgebra. J. Pure Appl. Algebra 216, 565–580 (2012)
Maruyama, Y.: From operational chu duality to coalgebraic quantum symmetry. In: Heckel, R., Milius, S. (eds.) CALCO 2013. LNCS, vol. 8089, pp. 220–235. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-40206-7_17
Maruyama, Y.: Duality theory and categorical universal logic: with emphasis on quantum structures. In: Proceedings of the Tenth Quantum Physics and Logic Conference, EPTCS, vol. 171, pp. 100–112 (2014)
Maruyama, Y.: AI, quantum information, and external semantic realism: Searle’s observer-relativity and Chinese room, revisited. In: Müller, V.C. (ed.) Fundamental Issues of Artificial Intelligence. SL, vol. 376, pp. 115–127. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-26485-1_8
Maruyama, Y.: Prior’s tonk, notions of logic, and levels of inconsistency: vindicating the pluralistic unity of science in the light of categorical logical positivism. Synthese 193, 3483–3495 (2016)
Maruyama, Y.: Meaning and duality: from categorical logic to quantum physics. DPhil Thesis, University of Oxford (2017)
Maruyama, Y.: quantum pancomputationalism and statistical data science: from symbolic to statistical AI, and to quantum AI. In: Müller, V.C. (ed.) PT-AI 2017. SAPERE, vol. 44, pp. 207–211. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-96448-5_20
Maruyama, Y.: Compositionality and contextuality: the symbolic and statistical theories of meaning. In: Bella, G., Bouquet, P. (eds.) CONTEXT 2019. LNCS (LNAI), vol. 11939, pp. 161–174. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-34974-5_14
Maruyama, Y.: The categorical integration of symbolic and statistical AI: quantum NLP and applications to cognitive and machine bias problems. In: Abraham, A., Siarry, P., Ma, K., Kaklauskas, A. (eds.) ISDA 2019. AISC, vol. 1181, pp. 466–476. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-49342-4_45
Maruyama, Y.: Post-truth AI and big data epistemology: from the genealogy of artificial intelligence to the nature of data science as a new kind of science. In: Abraham, A., Siarry, P., Ma, K., Kaklauskas, A. (eds.) ISDA 2019. AISC, vol. 1181, pp. 540–549. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-49342-4_52
Maruyama, Y.: Quantum physics and cognitive science from a Wittgensteinian perspective: Bohr’s Classicism, Chomsky’s Universalism, and Bell’s Contextualism. In: Wuppuluri, S., da Costa, N. (eds.) WITTGENSTEINIAN (adj.). TFC, pp. 375–407. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-27569-3_20
Maruyama, Y.: Symbolic and statistical theories of cognition: towards integrated artificial intelligence. In: Cleophas, L., Massink, M. (eds.) SEFM 2020. LNCS, vol. 12524, pp. 129–146. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-67220-1_11
Maruyama, Y.: Topological duality via maximal spectrum functor. Commun. Algebra 48, 2616–2623 (2020)
Maruyama, Y.: The conditions of artificial general intelligence: logic, autonomy, resilience, integrity, morality, emotion, embodiment, and embeddedness. In: Goertzel, B., Panov, A.I., Potapov, A., Yampolskiy, R. (eds.) AGI 2020. LNCS (LNAI), vol. 12177, pp. 242–251. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-52152-3_25
Maruyama, Y.: Category theory and foundations of life science. Biosytems 203, 104376 (2021)
Maruyama, Y.: Learning, development, and emergence of compositionality in natural language processing. In: Proceedings of IEEE ICDL, pp. 1–7 (2021)
Maruyama, Y.: Fibred algebraic semantics for a variety of non-classical first-order logics and topological logical translation. J. Symb. Logic 86, 1–27 (2021)
Minsky, M.L.: Logical versus analogical or symbolic versus connectionist or neat versus scruffy. AI Mag. 12, 34–51 (1991)
Norvig, P.: On chomsky and the two cultures of statistical learning. In: Pietsch, W., Ott, M. (eds.) Berechenbarkeit der Welt?, pp. 61–83. Springer, Wiesbaden (2017). https://doi.org/10.1007/978-3-658-12153-2_3
Selinger, P.: A survey of graphical languages for monoidal categories. In: Coecke, B. (ed.) New Structures for Physics, pp. 289–355. Springer, Heidelbrg (2011). https://doi.org/10.1007/978-3-642-12821-9_4
Sennesh, E.: Learning a Deep Generative Model like a Program: the Free Category Prior. arXiv:2011.11063 (2020)
Thompson, N., et al.: The computational limits of deep learning (2020). arXiv:2007.05558
Tshitoyan, V., et al.: Unsupervised word embeddings capture latent knowledge from materials science literature. Nature 571, 95–98 (2019)
Turney, P., Pantel, P.: From frequency to meaning: vector space models of semantics. J. Artif. Intell. Res. 37, 141–188 (2010)
Vicary, J.: The topology of quantum algorithms. In: Proceedings of 28th Annual ACM/IEEE Symposium on Logic in Computer Science, pp. 93–102 (2013)
Wittgenstein, L.: Philosophical Remarks. Blackwell, Hoboken (1975)
Xu, T., Maruyama, Y.: Neural String Diagrams: towards a universal compositional modelling language for categorical deep learning architectures. In: AGI (2021)
Zhang, Z., et al.: Deep learning on graphs: a survey. IEEE Trans. Knowl. Data Eng. 34, 249–270 (2022)
Acknowledgements
This work was supported by the Moonshot R&D Programme (JST; JPMJMS2033). Special thanks to Samson Abramsky and Bob Coecke.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Maruyama, Y. (2022). Categorical Artificial Intelligence: The Integration of Symbolic and Statistical AI for Verifiable, Ethical, and Trustworthy AI. In: Goertzel, B., Iklé, M., Potapov, A. (eds) Artificial General Intelligence. AGI 2021. Lecture Notes in Computer Science(), vol 13154. Springer, Cham. https://doi.org/10.1007/978-3-030-93758-4_14
Download citation
DOI: https://doi.org/10.1007/978-3-030-93758-4_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93757-7
Online ISBN: 978-3-030-93758-4
eBook Packages: Computer ScienceComputer Science (R0)