Formal Theory of Creativity, Fun, and Intrinsic Motivation (1990–2010)
The DOI matches its Crossref record; 125 of the 126 references it cites were found in a registry or an index.
Abstract
The simple, but general formal theory of fun and intrinsic motivation and creativity (1990-2010) is based on the concept of maximizing intrinsic reward for the active creation or discovery of novel, surprising patterns allowing for improved prediction or data compression. It generalizes the traditional field of active learning, and is related to old, but less formal ideas in aesthetics theory and developmental psychology. It has been argued that the theory explains many essential aspects of intelligence including autonomous development, science, art, music, and humor. This overview first describes theoretically optimal (but not necessarily practical) ways of implementing the basic computational principles on exploratory, intrinsically motivated agents or robots, encouraging them to provoke event sequences exhibiting previously unknown, but learnable algorithmic regularities. Emphasis is put on the importance of limited computational resources for online prediction and compression. Discrete and continuous time formulations are given. Previous practical, but nonoptimal implementations (1991, 1995, and 1997-2002) are reviewed, as well as several recent variants by others (2005-2010). A simplified typology addresses current confusion concerning the precise nature of intrinsic motivation.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“This underwrites Bayesian experimental design [134] and active learning in statistics [135], intrinsic motivation and artificial curiosity in machine learning and robotics [136, 137, 138, 139, 140].”
From The free energy principle made simpler but not too simple (2023) · cites this paper as [137]
In the citing paper
“This suggests that risk-averse policies minimise expected complexity or computational cost [137].”
From The free energy principle made simpler but not too simple (2023) · cites this paper as [137]
In the citing paper
“Ideally, the learner learns to choose actions that will lead to the largest improvements in its prediction or data compression performance (Schmidhuber, 2010).”
From Toward an integration of deep learning and neuroscience (2016) · cites this paper as Schmidhuber, 2010
In the citing paper
“In Schmidhuber (2010), this is done in the framework of reinforcement learning, and incorporates a mechanisms for the system to measure its own rate of learning.”
From Toward an integration of deep learning and neuroscience (2016) · cites this paper as Schmidhuber, 2010
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 7×
“Another potential drawback is the nature of the particular RNN-based RL method.”
BackgroundInfluentialSelf-citationA Mathematical Theory of Communication
Shannon, C. E. (1948) · Bell System Technical Journal · doi:10.1002/j.1538-7305.1948.tb01338.x · Cited by 88,486
Holland, J. H. (1992) · The MIT Press eBooks · doi:10.7551/mitpress/1090.001.0001 · Cited by 47,002 · cited 1×
“iii) An evolutionary algorithm [22], [29], [59],”
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225 · cited 7×
“Some Q-learning accelerators simply update pairs of actions and states with currently quickly changing Q-values more frequently than others (that is, Q-values with high first derivatives are favored).”
BackgroundMethodInfluentialReinforcement Learning: An Introduction
(2005) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2004.842673 · Cited by 24,496
Neural networks for pattern recognition
(1994) · Choice Reviews Online · doi:10.5860/choice.31-5500 · Cited by 18,538
Advances in neural information processing systems 7
(1996) · Computers & Mathematics with Applications · doi:10.1016/s0898-1221(96)90463-0 · Cited by 14,281
Neural Networks for Pattern Recognition
Bishop, C. M. (1995) · doi:10.1093/oso/9780198538493.001.0001 · Cited by 12,287
Watkins, C. J. C. H., & Dayan, P. (1992) · Machine Learning · doi:10.1007/bf00992698 · Cited by 9,083
Reinforcement Learning: A Survey
Kaelbling, L. P., Littman, M. L., & Moore, A. W. (1996) · Journal of Artificial Intelligence Research · doi:10.1613/jair.301 · Cited by 8,972 · cited 4×
“, some typically suboptimal RL method [35].”
BackgroundMethodInfluential
Research works printed but not linked: 24
- kullback, 1959, Statistics and Information Theory
- hutter, 2004, Universal Artificial Intelligence Sequential Decisions Based on Algorithmic Probability
- holland, 1975, Adaptation in Natural and Artificial Systems
- schmidhuber, 2010, How to Learn Programs with Artificial Recurrent Neural Networks
- harlow, 1950, novelty and curiosity as determinants of exploratory behavior, J Exp Psychol, 41, 68
- berlyne, 1950, novelty and curiosity as determinants of exploratory behavior, Brit J Psychol, 41, 68
- 10.1007/bf00993591 · doi:10.1007/bf00993591 · Paper page
- levin, 1973, universal sequential search problems, Problems Inform Trans, 9, 265
- schmidhuber, 1991, Advances in Neural Information Processing Systems 3, nips 3, 500
- schmidhuber, 1997, Femmes Fractales
- schmidhuber, 2002, Advances in Evolutionary Computing, 579
- ring, 1994, Continual Learning in Reinforcement Environments
- schmidhuber, 1990, Dynamische neuronale Netze und das fundamentale raumzeitliche Lernproblem
- schmidhuber, 1991, Adaptive Curiosity and Adaptive Confidence
- schmidhuber, 2009, Compression Progress The Algorithmic Principle Behind Curiosity and Creativity (With Applications of the Theory of Humor) 40 min Video of Invited Talk at Singularity Summit 2009
- schmidhuber, 2009, Multiple Ways to Design Research Research Cases that Reshape the Design Discipline, 98
- singh, 2005, Advances in neural information processing systems, 17
- schwefel, 1974, Numerische Optimierung von Computer-Modellen
- schmidhuber, 2007, alle berechenbaren universen (all computable universes), Spektrum der Wissenschaft, 75
- storck, 1995, reinforcement driven information acquisition in non-deterministic environments, Proc Int Conf Artif Neural Netw, 2, 159
- brafman, 2002, R-MAX—A general polynomial time algorithm for near-optimal reinforcement learning, J Mach Learn Res, 3, 213
- frank, 2002, sthetische Information Estetika informacio Eine Einfhrung in die Kybernetische sthetik
- franke, 1979, Kybernetische sthetik Phnomen Kunst
- schmidhuber, 2006, Artificial General Intelligence, 175
Cited by
Curiosity-Driven Exploration by Self-Supervised Prediction
Pathak, D., Agrawal, P., Efros, A. A., & Darrell, T. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2017.70 · Cited by 1,815
“Note that ICM-pixels is representative of previous methods which compute information gain by directly using the observation space (Schmidhuber, 2010; Stadie et al., 2015).”
BackgroundMethodInfluentialAn Introduction to Deep Reinforcement Learning
Vincent, F.-L., Peter, H., Riashat, I., Marc, G. B., & Joelle, P. (2018) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000071 · Cited by 1,286
“Another common approach is to have a directed scheme thanks to exploration rewards given to the agent via heuristics that estimate novelty (Schmidhuber, 2010; Stadie et al., 2015; Houthooft et al., 2016).”
REBUS and the anarchic brain: Toward a unified model of the brain action of psychedelics
Carhart-Harris, R. L., & Friston, K. J. (2019) · Pharmacological Reviews · doi:10.1124/pr.118.017160 · Cited by 1,180
“, 2017) and is consistent with an explorative search strategy, which entails exploring novelty for the sake of significant knowledge gain (also known as intrinsic motivation in robotics) (Oudeyer and Kaplan, 2007; Schmidhuber, 2010; Barto et al., 2013).”
BackgroundMulti-agent deep reinforcement learning: a survey
Gronauer, S., & Diepold, K. (2022) · Artificial Intelligence Review · doi:10.1007/s10462-021-09996-w · Cited by 957
“One way to realize this is with intrinsic motivation, which is based on the concept of maximizing an internal reinforcement signal by actively discovering novel or surprising patterns (Chentanez et al. 2005; Oudeyer and Kaplan 2007; Schmidhuber 2010).”
BackgroundHierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
Kulkarni, T. D., Narasimhan, K., Saeedi, A., & Tenenbaum, J. B. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1604.06057 · Cited by 770
Metadata
- Authors
- Schmidhuber, J.
- Affiliations
- Dalle Molle Institute for Artificial Intelligence Research (CH)
- Corresponding author
- Schmidhuber, Jürgen
- ORCID
- Schmidhuber, Jürgen
- Venue
- IEEE Transactions on Autonomous Mental Development
- Published
- 2010-09
- Volume
- 2
- Issue
- 3
- Pages
- 230-247
- ISSN
- 1943-0604 (ISSN-L)
- 1943-0612
- OpenAlex
- W2034806191
- Cited by
- 747
- References
- 126
- Citations by year
- 2026: 29 · 2025: 64 · 2024: 46 · 2023: 50 · 2022: 60 · 2021: 69 · 2020: 92 · 2019: 72 · 2018: 55 · 2017: 37 · 2016: 26 · 2015: 24 · 2014: 30 · 2013: 28 · 2012: 41
- FWCI
- 31.09
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Transactions on Autonomous Mental Development (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Artificial Intelligence › Psychiatry, Mental Health, Neuroscience
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Creativity in Education and Neuroscience
- Keywords
- intrinsic motivation, creativity, fun, intrinsic rewards, active learning (machine learning), novelty seeking, surprise, pattern discovery, data compression, online prediction, autonomous agents, robots, aesthetics, humor, intelligence, limited computational resources, continuous-time formulation, discrete-time formulation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.