Abstract
Any machine targeted for human-level intelligence must be able to autonomously use its prior experience in novel situations, unforeseen by its designers. Such knowledge transfer capabilities are usually investigated under an assumption that a learner receives training in a source task and is subsequently tested on another similar target task. However, most current AI approaches rely heavily on human programmers, who choose these tasks based on their intuition. Another largely unaddressed approach is to provide an artificial agent with methods for transferring relevant knowledge autonomously. One step towards effective autonomous generalization capabilities builds on (autonomous) causal modeling and inference processes, using task-independent knowledge representations. We describe a controller that enables an agent to intervene on a dynamical task to discover and learn its causal relations cumulatively from experience. Our controller bootstraps its learning from knowledge of correlation, then removes non-direct-cause correlations – correlations that are due to a common (external) cause, be spurious, or invert cause and effect – through strategic causal interventions, while learning the functions relating a task’s causal variables. The effectiveness of knowledge transfer by the proposed controller is tested through simulation experiments.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
By ‘general solution’ we mean that the learning is largely independent of the task-environment and can be used to transfer learned skills between different task types.
- 2.
This is a special case of the ‘Independence of Cause and Mechanism’ principle, which states that the mechanism that connects the cause to the effect is independent of the cause itself; i.e. X causes Y if and only if \(P(Y\mid X)\) is independent of P(X) [12].
- 3.
Our physical formalization is compatible with event-based causality, where an event causes another event to happen. An event can be defined as a set of manipulable variables with changing values in a time interval that apply forces and causes changes in values of another set of variables in a subsequent time interval.
References
Baumann, D., Solowjow, F., Johansson, K.H., Trimpe, S.: Identifying causal structure in dynamical systems. arXiv preprint arXiv:2006.03906 (2020)
Bouvier, V., Very, P., Hudelot, C., Chastagnol, C.: Hidden covariate shift: a minimal assumption for domain adaptation. Technical report, arXiv preprint arXiv:1907.12299 (2019)
Drescher, G.L.: Made-Up Minds: A Constructivist Approach to Artificial Intelligence. MIT Press (1991)
Ke, Z., Li, Z., Cao, Z., Liu, P.: Enhancing transferability of deep reinforcement learning-based variable speed limit control using transfer learning. IEEE Trans. Intell. Transp. Syst. 22, 4684–4695 (2020)
Nivel, E., et al.: Bounded recursive self-improvement. arXiv preprint arXiv:1312.6764 (2013)
Pearl, J.: Causality, pp. 22–24. Cambridge University Press (2009)
Pearl, J.: Theoretical impediments to machine learning with seven sparks from the causal revolution. arXiv preprint arXiv:1801.04016 (2018)
Peters, J., Bühlmann, P., Meinshausen, N.: Causal inference by using invariant prediction: identification and confidence intervals. J. Roy. Stat. Soc. Ser. B (Stat. Methodol.) 78, 947–1012 (2016)
Peters, J., Janzing, D., Schölkopf, B.: Elements of Causal Inference: Foundations and Learning Algorithms, pp. 15–26, 88. The MIT Press (2017)
Piaget, J., Piercy, M., Berlyne, D.: The Psychology of Intelligence (1951)
Rojas-Carulla, M., Schölkopf, B., Turner, R., Peters, J.: Invariant models for causal transfer learning. Int. J. Biostat. 19(1), 1309–1342 (2018)
Shajarisales, N., Janzing, D., Schölkopf, B., Besserve, M.: Telling cause from effect in deterministic linear dynamical systems. In: International Conference on Machine Learning, pp. 285–294. PMLR (2015)
Sheikhlar, A., Thórisson, K.R., Eberding, L.M.: Autonomous cumulative transfer learning. In: Goertzel, B., Panov, A.I., Potapov, A., Yampolskiy, R. (eds.) AGI 2020. LNCS (LNAI), vol. 12177, pp. 306–316. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-52152-3_32
Tan, C., Sun, F., Kong, T., Zhang, W., Yang, C., Liu, C.: A survey on deep transfer learning. In: Kůrková, V., Manolopoulos, Y., Hammer, B., Iliadis, L., Maglogiannis, I. (eds.) ICANN 2018. LNCS, vol. 11141, pp. 270–279. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01424-7_27
Taylor, M.E., Stone, P.: Transfer learning for reinforcement learning domains: a survey. J. Mach. Learn. Res. 10(7), 1633–1685 (2009)
Thórisson, K.R.: A new constructivist AI: from manual methods to self-constructive systems. In: Wang, P., Goertzel, B. (eds.) Theoretical Foundations of Artificial General Intelligence. Atlantis Thinking Machines, vol. 4. Atlantis Press, Paris (2012). https://doi.org/10.2991/978-94-91216-62-6_9
Thórisson, K.R.: Seed-programmed autonomous general learning. Proc. Mach. Learn. Res. 131, 32–70 (2020)
Thórisson, K.R., Bieger, J., Li, X., Wang, P.: Cumulative learning. In: Proceedings of the 12th International Conference on Artificial General Intelligence, pp. 198–208 (2019)
Thórisson, K.R., Talbot, A.: Cumulative learning with causal-relational models. In: Iklé, M., Franz, A., Rzepka, R., Goertzel, B. (eds.) AGI 2018. LNCS (LNAI), vol. 10999, pp. 227–237. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-97676-1_22
Wang, P.: Rigid Flexibility: The Logic of Intelligence. Springer, Dordrecht (2006). https://doi.org/10.1007/1-4020-5045-3
Acknowledgments
This work was supported in part by Cisco Systems, the Icelandic Institute for Intelligent Machines and Reykjavik University.
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
Sheikhlar, A., Eberding, L.M., Thórisson, K.R. (2022). Causal Generalization in Autonomous Learning Controllers. 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_24
Download citation
DOI: https://doi.org/10.1007/978-3-030-93758-4_24
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)