Megerősítéses tanulás a pszichológiában és a mesterséges intelligenciában

Levente Zsolt Juhász

doi:10.63684/dk.2026.09.03

Levente Zsolt Juhász Dunaújvárosi Egyetem, Tanárképző Központ, egyetemi docens https://orcid.org/0009-0006-0240-2937

DOI: https://doi.org/10.63684/dk.2026.09.03

Keywords: Reinforcement learning, conditioning, artificial intelligence

Abstract

Reinforcement learning (RL) is the scientific study of how animals and machines can shape their behavior to maximize rewards. The roots of RL research go back to early work in psychology on instrumental learning.

Modern RL, however, is a highly interdisciplinary field that lies at the intersection of computer science, machine learning, psychology, and neuroscience. The presentation will summarize the most important mathematical principles of the field, such as the exploration/exploitation dilemma, temporal-difference learning (TD-learning), Q-learning, and the differences between model-based and model-free learning. Finally, it will review a number of open questions both in psychology and neuroscience.

References

Smith, E. E.–Nolen-Hoeksema, S.–Fredrickson, B. L.–Loftus, G. R. (2005): Atkinson–Hilgard: Pszichológia. Budapest: Osiris.

Gluck, M. A.–Mercado, E.–Myers, C. E. (2019): Learning and Memory: From Brain to Behavior. New York: Worth Publishers.

Domjan, M.–Delamater, A. R. (2023): The Essentials of Conditioning and Learning. New York: American Psychological Association.

Mitchell, T. (1997): Machine Learning. New York: McGraw-Hill.

Géron, A. (2026): Hands-On Machine Learning with Scikit- Learn and PyTorch. New York: O’Reilly.

Sutton, R. S.–Barto, A. G. (2018): Reinforcement Learning. New York: Bradford Books.

Mnih, V. e. (2015): Human-level control through deep reinforcement. Nature, 518., pp. 529–533.

Silver, D. E. (2016): Mastering the game of Go with deep neural networks and tree search. Nature, 529., pp. 484–489.

Silver D, E. A. (2017): Mastering the game of Go without human knowledge. Nature, 550., pp. 354-359.

Masset, P.–Geshman, S. J. (2025): Reinforcement learning with dopamine: A convergence of natural and artificial intelligence. In: S. J. Cragg,–M. E. Walton: The Handbook of Dopamine. pp. 305–318. New York: Academic Press.

Seriés, P. (2020): Computational Psychiatry. Boston: The MIT Press.

Heinz, A. (2017): A New Understanding of Mental Disorders. Boston: The MIT Press.

Murphy, K. P. (2026): Reinforcement Learning: An Overview.

Norvig, P.–Russell, S. (2023): Mesterséges intelligencia 1–2. Budapest: Taramix.

Chollet, F. (2026): Deep Learning with Python. New York: Manning.

LeCun, Y. (2025): A mesterséges intelligencia és a mélytanulás forradalma. Budapest: TypoTex.