Abstract
An ADAM artificial psyche model implementing a hierarchical deep reinforcement learning architecture is proposed. ADAM is able to learn increasingly complex and time-consuming behavioral skills as the number of artificial psyche control levels increases. Purposeful behavior is formed by a hierarchical learning system with a gradual increase in the number of levels, where each hierarchical level is responsible for its own time scale of behavior.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig1_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig2_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig3_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig4_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig5_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig6_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig7_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig8_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1134%2FS0005117922060030/MediaObjects/10513_2022_2286_Fig9_HTML.jpg)
Similar content being viewed by others
REFERENCES
Russell, S. and Norvig, P., Artificial Intelligence: A Modern Approach, Harlow: Pearson, 2009.
Nikolenko, S., Kadurin, A., and Arkhangel’skaya, E., Glubokoe obuchenie. Pogruzhenie v mir neironnykh setei (Deep Learning. Dive into the World of Neural Networks), St. Petersburg: Piter, 2018.
Bengio, Y., From system-1 deep learning to system-2 deep learning, Thirty-Third Conf. Neural Inf. Process. Syst., 2019.
Kotseruba, I. and Tsotsos, J.K., 40 years of cognitive architectures: core cognitive abilities and practical applications, Artif. Intell. Rev., 2020, vol. 53, no. 1, pp. 17–94.
Silver, D., Singh, S., Precup, D., and Sutton, R.S., Reward is enough, Artif. Intell., 2021, article ID 103535.
Laird, J.E., Lebiere, C., and Rosenbloom, P.S., A standard model of the mind: toward a common computational framework across artificial intelligence, cognitive science, neuroscience, and robotics, AI Mag., 2017, vol. 38, no. 4, pp. 13–26.
Russell, S., Human Compatible: Artificial Intelligence and the Problem of Control, Viking, 2019.
Silver, D. et al., Mastering chess and shogi by self-play with a general reinforcement learning algorithm. .
Shumskii, S.A., Deep structural learning: a new look at reinforcement learning, Sb. nauchn. tr. XX Vseross. nauchn. konf. “Neiroinformatika-2018. Lektsii po neiroinformatike” (Collect. Sci. Pap. XX All-Russ. Sci. Conf. “Neuroinformatics-2018. Lectures on Neuroinformatics”) (Moscow, 2018), pp. 11–43.
Shumskii, S.A., Reengineering of brain architecture: the role and interaction of the main subsystems, Sb. nauchn. tr. XX Vseross. nauchn. konf. “Neiroinformatika-2018. Lektsii po neiroinformatike” (Collect. Sci. Pap. XX All-Russ. Sci. Conf. “Neuroinformatics-2015. Lectures on Neuroinformatics”) (Moscow, 2015), pp. 13–45.
Clark, A., Surfing Uncertainty: Prediction, Action, and the Embodied Mind, Oxford: Oxford Univ. Press, 2015.
Friston, K.J., Waves of prediction, PLoS Biol., 2019, vol. 17, no. 10, p. e3000426.
Mountcastle, V.B., The columnar organization of the neocortex, Brain: J. Neurology, 1997, vol. 120, no. 4, pp. 701–722.
Kohonen, T., Self-organized formation of topologically correct feature maps, Biol. Cybern., 1982, vol. 43, no. 1, pp. 59–69.
Sutton, R.S. and Barto, A.G., Reinforcement Learning: an Introduction, Boston: MIT Press, 2018.
Shumskii, S.A. and Baskov, O.V., ADAM Deep Control Deep hierarchical reinforcement software agent, State Regist. Comput. Programs, reg. no. RU 2021660307.
Funding
This work was supported in part by the Competence Center of the National Technological Initiative in the field of “Artificial Intelligence” at the Moscow Institute of Physics and Technology.
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated by V. Potapchouck
Rights and permissions
About this article
Cite this article
Shumskii, S.A. ADAM: a Model of Artificial Psyche. Autom Remote Control 83, 847–856 (2022). https://doi.org/10.1134/S0005117922060030
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0005117922060030