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
Items 601–700 of 886
Lian, R., & Goertzel, B. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-00686-8_35 · Cited by 2
Cheng, J., Li, N., Dong, C., & Tang, C. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114534 · Cited by 3
From optimization to inquiry: a Deweyan criterion for machine intelligence
ABUKAR, G. (2025) · AI and Ethics · doi:10.1007/s43681-025-00911-5 · Cited by 2
“Recent work in active inference robotics [12, 13], intrinsic motivation systems [14], and deliberative AI architectures [15] provides concrete technical frameworks that align with the Deweyan model, demonstrating that problem-space reconstruction is not merely philosophical speculation but an…”
Machine Learning for Microbiome Statistics
Xia, Y., & Sun, J. (2026) · Chapman & Hall/CRC eBooks · doi:10.1201/9781003610281 · Cited by 3
The authenticity gap: Why generative AI cannot replicate human creativity
Kharkhurin, A. V., & Corazza, G. E. (2026) · Journal of Creativity · doi:10.1016/j.yjoc.2026.100132 · Cited by 3
The Pleasure of Habituation: Aesthetic Appreciation between Familiarity and Novelty
Bertinetto, A., & Frascaroli, J. (2026) · The British Journal of Aesthetics · doi:10.1093/aesthj/ayag030 · Cited by 2
Active Learning for Player Modeling
Shaker, N., Abou-Zleikha, M., & Shaker, M. (2015) · Foundations of Digital Games · Cited by 3
Efficient and Scalable Exploration via Estimation-Error
Sun, C., Wang, R., Li, R., Wu, J., & Hu, X. (2019) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2019.8852234 · Cited by 2
Nagashima, K., Morita, J., & Takeuchi, Y. (2021) · Transactions of the Japanese Society for Artificial Intelligence · doi:10.1527/tjsai.36-5_ag21-e · Cited by 2
A Learnable Noise Exploration Method for Multi-Agent Reinforcement Learning
Zhao, Z., Zhang, Y., Wang, S., & Qu, H. (2023) · International Annual Conference on Complex Systems and Intelligent Science (CSIS-IAC) · doi:10.1109/csis-iac60628.2023.10363845 · Cited by 2
“Similarly, compression progress [16] and variational information maximization [17], the former through online prediction and compression of simplified intrinsic rewards, and the latter utilizing variational inference to solve exploration in continuous state and action spaces.”
BackgroundLearning Adaptable Utility Models for Morphological Diversity
Campos-Alfaro, F., Jara, C., Romero, A., Naya-Varela, M., & Duro, R. J. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-61137-7_11 · Cited by 2
Interactive and Explainable Robot Learning: A Comprehensive Review
Seraj, E., Lee, K. M., Zaidi, Z., Xiao, Q., Li, Z., Nascimento, A., van Waveren, S., Tambwekar, P., Paleja, R., Das, D., et al. (2024) · Foundations and Trends in Robotics · doi:10.1561/2300000081 · Cited by 7
Open-Vocabulary Predictive World Models from Sensor Observations
Karlsson, R., Asfandiyarov, R., Carballo, A., Fujii, K., Ohtani, K., & Takeda, K. (2024) · Sensors · doi:10.3390/s24144735 · Cited by 2
“The idea of using machine learning to learn predictive models of the world was proposed by Schmidhuber [3–5].”
MethodAutonomous Mental Development at the Individual and Collective Levels: Concept and Challenges
Lippi, M., Mariani, S., Martinelli, M., & Zambonelli, F. (2024) · IEEE Access · doi:10.1109/access.2024.3522362 · Cited by 2
“Reinforcement learning approaches based on intrinsic rewards [73], instead, more closely exploit the idea of exploring the world to develop a sense of agency.”
Innate-Values-Driven Reinforcement Learning Based Cognitive Modeling
Yang, Q. (2025) · IEEE Conference on Cognitive and Computational Aspects of Situation Management (CogSIMA) · doi:10.1109/cogsima64436.2025.11079473 · Cited by 2
An Anti-hebbian Learning Rule to Represent Drive Motivations for Reinforcement Learning
Kompella, V. R., Kazerounian, S., & Schmidhuber, J. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-08864-8_17 · Cited by 2
“The theory of Artificial Curiosity (AC; [21]) introduces a mathematical formalism for describing curiosity and creativity in artificial agents.”
BackgroundSelf-citationExploring the Periphery of Knowledge by Intrinsically Motivated Systems
Makukhin, K., & Bolland, S. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-14803-8_4 · Cited by 2
“In contrast, the field of developmental robotics has recently made substantial progress in modelling intrinsic motivation and applying it to artificial systems, see [5, 6] for reviews.”
BackgroundSchneider, M., & Adamy, J. (2015) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2015.7280623 · Cited by 2
The perceptions of creativity and creative potentials of future kindergarten and class teachers
Milovanović, R., & Kopas-Vukašinović, E. M. (2014) · Zbornik Instituta za pedagoska istrazivanja · doi:10.2298/zipi1401181m · Cited by 2
A Multiple Context Brain for Experiments With Robot Consciousness
Andreae, J. H. (2011) · IEEE Transactions on Autonomous Mental Development · doi:10.1109/tamd.2011.2164404 · Cited by 2
“They are embarrassingly simple compared with Schmidhuber's [22] efforts to measure the improvements of the brain's world model.”
BackgroundKnowing When to Abandon Unproductive Learning
Shultz, T. R., Doty, E., & Dandurand, F. (2012) · eScholarship (California Digital Library) · Cited by 2
Branković, S. (2013) · PubMed · Cited by 3
On the use of intrinsic motivation for visual saliency learning
Craye, C., Filliat, D., & Goudou, J.-F. (2016) · IEEE International Conference on Development and Learning (ICDL) · doi:10.1109/devlrn.2016.7846811 · Cited by 2
“To overcome limitations related to novelty or error in unlearnable situations, intrinsic motivation based on progress has been proposed [2], [20], [21].”
BackgroundFactors that Influence the Creativity of Engineering Students
Genco, N., Hölttä‐Otto, K., & Seepersad, C. C. (2020) · ASEE Annual Conference & Exposition · doi:10.18260/1-2--17986 · Cited by 2
Q-map: a Convolutional Approach for Goal-Oriented Reinforcement Learning
Pardo, F., Levdik, V., & Kormushev, P. S. (2018) · Spiral (Imperial College London) · Cited by 6
On the Effects of Simulating Human Decisions in Game Analysis
Volz, V., & Naujoks, B. (2019) · IEEE Conference on Games (CoG) · doi:10.1109/cig.2019.8848071 · Cited by 2
Assessing Aesthetics of Generated Abstract Images Using Correlation Structure
Khajehabdollahi, S., Martius, G., & Levina, A. (2019) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci44817.2019.9002779 · Cited by 3
An Interruptible Task Allocation Model: Application to a Honey Bee Colony Simulation
Alves, T., Rivière, J., Alaux, C., Conte, Y. L., Singhoff, F., Duval, T., & Rodin, V. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-49778-1_1 · Cited by 2
Upgrading Gestalt psychology with variational neuroethology: The case of perceptual pleasures
de Cruys, S. V. (2018) · Physics of Life Reviews · doi:10.1016/j.plrev.2018.01.002 · Cited by 4
Novelty-Guided Reinforcement Learning via Encoded Behaviors
Ramamurthy, R., Sifa, R., Lübbering, M., & Bauckhage, C. (2020) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9206982 · Cited by 3
“Early work [27] in this regard proposed to learn a curiosity model which predicts future events using history of interactions with environment.”
BackgroundMühlenhoff, M., & Rudloff, D. (2020) · Springer eBooks · doi:10.1007/978-3-658-28973-7_21-1 · Cited by 2
Self-Development and Causality in Intelligent Environments
Martinelli, M., Mariani, S., Lippi, M., & Zambonelli, F. (2022) · Ambient intelligence and smart environments · doi:10.3233/aise220050 · Cited by 2
Autonomous Exploration for Navigating in MDPs Using Blackbox RL Algorithms
Gajane, P., Auer, P. L., & Ortner, R. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/413 · Cited by 2
“…curiosity-driven learning, automatic goal generation etc. Owing to the space restrictions, a necessarily incomplete list of these works include [Schmidhuber, 2010; Singh et al. , 2004; Singh et al. , 2010; Oudeyer and Ka-plan, 2007; Baranes and Oudeyer, 2009; Lopes et al. , 2012; Gottlieb et…”
BackgroundUsing Curiosity for an Even Representation of Tasks in Continual Offline Reinforcement Learning
Pathmanathan, P., Díaz-Rodríguez, N., & Ser, J. D. (2023) · Cognitive Computation · doi:10.1007/s12559-023-10213-9 · Cited by 3
“This can broadly be categorized as those who measure the novelty of a state [40, 41] or those who measure the uncertainty of the agent in predicting the consequences of its actions [42, 43].”
Zanzarella, I. (2025) · PhilPapers (PhilPapers Foundation) · Cited by 2
Computational Curiosity (A Book Draft)Preprint
Wu, Q. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1502.04780 · Cited by 2
Hierarchical Optimistic Region Selection driven by CuriosityPreprint
Maillard, O.-A. (2012) · HAL · Cited by 2
Deep Curiosity Loops in Social EnvironmentsPreprint
Barkan, J., & Gordon, G. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.03645 · Cited by 2
“Curiosity has been described as a predominant driver of human behavior and it had not gone unnoticed by the Artificial Intelligence community [29, 30].”
BackgroundA Definition of Happiness for Reinforcement Learning AgentsPreprint
Daswani, M., & Leike, J. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1505.04497 · Cited by 2
“Schmidhuber’s work on ‘intrinsic motivation’ adds a related component to the reward in order to motivate the agent to explore in interesting directions [9].”
BackgroundAutonomous Reinforcement of Behavioral Sequences in Neural DynamicsPreprint
Kazerounian, S., Luciw, M., Richter, M., & Sandamirskaya, Y. (2012) · arXiv (Cornell University) · doi:10.48550/arxiv.1210.3569 · Cited by 2
“One could imagine future work using more sophisticated exploration methods [26].”
MethodLearning the Arrow of TimePreprint
Rahaman, N., Wolf, S., Goyal, A., Remme, R., & Bengio, Y. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.01285 · Cited by 4
“One typical approach towards encouraging curious behaviour is to seek novel states that surprise the agent (Schmidhuber, 2010; Pathak et al., 2017; Burda et al., 2018) and use the error in the agent’s prediction of future states is used as a curiosity reward.”
BackgroundMeta Adaptation using Importance Weighted DemonstrationsPreprint
Lekkala, K. K., Abu-El-Haija, S., & Itti, L. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.10322 · Cited by 2
“We corroborate this hypothesis with previous works in Neuroscience such as [31, 32].”
ResultAutonomous learning of multiple, context-dependent tasksPreprint
Santucci, V. G., Montella, D., da Silva, B. C., & Baldassarre, G. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.13847 · Cited by 2
Active Inference as a Framework for Brain-Computer InterfacesPreprint
Ather, S. H. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.02.02.429272 · Cited by 2
“Schmidhuber put forward a simple formal theory of fun and intrinsic motivation based on maximizing intrin-222 sic reward for active creation or discovery of novel, surprising patterns[15].”
BackgroundFriston, K. J. (2021) · Journal of Consciousness Studies · doi:10.53765/20512201.28.11.178 · Cited by 2
Zhang, S., Tian, Y., Liu, Q., & Wu, H. (2024) · eLife · doi:10.7554/elife.92892.2 · Cited by 2
Merits of curiosity: a simulation studyPreprint
Gruaz, L., Modirshanechi, A., Becker, S., & Brea, J. (2024) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/evm9n · Cited by 5
Emergence of Implicit World Models from Mortal AgentsPreprint
Horibe, K., & Yoshida, N. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2411.12304 · Cited by 2
The intrinsic motivation of reinforcement and imitation learning for sequential tasksPreprint
Nguyen, S. M. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2412.20573 · Cited by 2
Subjective functionsPreprint
Gershman, S. J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.15948 · Cited by 2
“A related line of work in model-based systems uses the unsigned error of model predictions to guide exploration and goal selection (e.g., Schmidhuber, 1991, 2010; Oudeyer et al., 2007; Pathak et al., 2017; Molinaro et al., 2024).”
Active physical inference via reinforcement learningPreprint
Li, S., Sun, Y., Liu, S., Sun, Y., Gureckis, T. M., & Bramley, N. R. (2019) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/6vr4g · Cited by 2
“The idea that the brain seeks information as a form of intrinsic reward captures a middle ground idea that that for humans, discovery is an “end in itself” driven by something we intuitively understand as “curiosity” (Gottlieb, Oudeyer, Lopes, & Baranes, 2013; Schmidhuber, 2010).”
BackgroundIntrinsic exploration for reinforcement learning beyond rewards
Castanyer, R. C. (2024) · Papyrus : Institutional Repository (Université de Montréal) · doi:10.71781/10181 · Cited by 2
The Information-Theoretic Imperative: Compression and the Epistemic Foundations of IntelligencePreprint
Dittrich, C., & Kinne, J. F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2510.25883 · Cited by 2
“But when confronted with out-of-distribution inputs that lack this evolutionary filtering, the purely statistical shortcuts can fail, and the system’s lack of direct grounding becomes apparent (Schmidhuber, 2010; Amodei et al., 2016).”
InfluentialDéjàQ: Open-Ended Evolution of Diverse, Learnable and Verifiable ProblemsPreprint
Röpke, W., Coward, S., Lupu, A., Foster, T., Rocktäschel, T., & Foerster, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2601.01931 · Cited by 2
“Related ideas appear in the intrinsic-motivation and curiosity literature, where learning progress quantifies improvements in prediction or goal achievement and is used to build automatic curricula (Oudeyer et al., 2007; Schmidhuber, 2010).”
Cognitive Training for Language Models: Towards General Capabilities via Cross-Entropy GamesPreprint
Hongler, C., Gabriel, F., Hartmann, V. N., Renard, A., & Emil, A. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.22479 · Cited by 2
Construction sociale d'une esthétique artificielle : Berenson, un robot amateur d'art
Karaouzene, A. (2017) · theses.fr (ABES) · doi:10.70675/f458a4dez4d3ez4256zaae4z55e83c03479a · Cited by 2
Gaya, M. C. (2017) · theses.fr (ABES) · doi:10.70675/50afa065z5b0cz41d7z8e69zc8a16f165b1d · Cited by 3
“Interested readers on various typological classifications of computational motivational systems are referred to [13, 91, 90, 107, 82, 57].”
BackgroundPlanning to Explore: Curiosity-Driven Planning for LLM Test GenerationPreprint
Amayuelas, A., Laakom, F., Piękos, P., Wang, W., Xu, Y., Wang, Y., Schmidhuber, J., & Wang, W. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.05159 · Cited by 2
“In exploration theory, Schmidhuber (1991b; 2010) formalized curiosity as the drive to maximize learning progress by seeking experiences that improve the agent’s world model.”
Self-citationEmergence of Exploration in Policy Gradient Reinforcement Learning via RetryingPreprint
Nishimori, S., Parmas, P., Koyamada, S., Kozuno, T., Kitamura, T., Ishii, S., & Matsuo, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.00151 · Cited by 2
“Information-gain-based methods instead seek states that reduce model uncertainty (Schmidhuber, 2010; Sun et al., 2011; Houthooft et al., 2016; Sukhija et al., 2025).”
Retry Policy Gradients in Continuous Action SpacesPreprint
Nishimori, S., & Parmas, P. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.05888 · Cited by 2
“…include count-based exploration (Bellemare et al., 2016; Ostrovski et al., 2017; Lobel et al., 2023; Tang et al., 2017), prediction-error bonuses (Burda et al., 2019; Pathak et al., 2017), and information gain or curiosity signals (Houthooft et al., 2016; Sukhija et al., 2025; Schmidhuber, 2010).”
Experimental Evidence that Empowerment May Drive Exploration in Sparse-Reward Environments
Massari, F., Biehl, M., Meeden, L., & Kanai, R. (2021) · IEEE International Conference on Development and Learning (ICDL) · doi:10.1109/icdl49984.2021.9515647 · Cited by 5
Holm, L., & Schrater, P. (2024) · Behavioral and Brain Sciences · doi:10.1017/s0140525x23003321 · Cited by 1
Curious Feature Selection-Based Clustering
Moran, M., & Gordon, G. (2024) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2024.3407034 · Cited by 1
Contrast Information Dynamics: A Novel Information Measure for Cognitive Modelling
Homer, S. T., Harley, N., & Wiggins, G. A. (2024) · Entropy · doi:10.3390/e26080638 · Cited by 2
Spiekermann, S. (2024) · Springer eBooks · doi:10.1007/978-3-658-19606-6_35 · Cited by 2
Intrinsically motivated norm compliance and the sense of obligation
Michael, J., & Tummolini, L. (2025) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2025.102043 · Cited by 5
Leveraging psychedelic neuroscience to boost human creativity using artificial intelligence
Ross, B. M. (2025) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2025.1589086 · Cited by 1
“Moreover, incorporating emotional-responses, such as surprise, pleasure, or esthetic dissonance, may help prioritize outputs that more closely mirror the psychological richness of psychedelic insights (Kuypers et al., 2016; Schmidhuber, 2010; Maher et al., 2013).”
Intrinsic Motivation for Artificial Agents
Robertson, P., & Georgeon, O. L. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-96325-4_5 · Cited by 1
Synthesizing Evolving Symbolic Representations for Autonomous Systems
Sartor, G., Oddi, A., Rasconi, R., Santucci, V. G., & Meo, R. (2025) · Progress in Artificial Intelligence · doi:10.1007/s13748-025-00394-9 · Cited by 1
“Indeed, some models have been implemented using the prediction error (PE) in anticipating the effect of agent’s actions (and more precisely the improvement of the prediction error [64, 65]) as an IM signal.”
Parallel Fitness Scores Evaluation to Improve Training Speed of the NEAT Algorithm Using GO Routines
Omelianenko, I. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-6938-7_17 · Cited by 1
Robust inference via top-down feedback and neural variability in recurrent vision models
Greco, A., D’Alessandro, M., Friston, K. J., Pezzulo, G., & Siegel, M. (2026) · Patterns · doi:10.1016/j.patter.2026.101693 · Cited by 2
Efficient coding in active perception: A developmental perspective on autonomous control
López, F. M., Shi, B. E., & Triesch, J. (2026) · Advances in child development and behavior · doi:10.1016/bs.acdb.2026.04.003 · Cited by 2
Supporting Decision-Making for City Management Through Automated Planning and Execution
Benedictis, R. D., Beraldo, G., Cesta, A., & Cortellessa, G. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-80607-0_10 · Cited by 1
“The higher-level tasks (a.k.a. intrinsic motivations [21]) generated by the deliberative tier during plan execution are only one factor influencing User-defined function Description new_solver(purpose, files) Creates a new solver for the given purpose and starts solving the planning problem…”
Reviewing, Creativity, and Algorithmic Information Theory
Brown, D., & Peeperkorn, M. (2023) · Kent Academic Repository (University of Kent) · Cited by 1
Novel Goal Creation and Evaluation in Open-Ended Games
Cognitive Science Society 2025, Chu, J., Davidson, G., Todd, G., & Xu, W. (2025) · eScholarship (California Digital Library) · doi:10.48448/53g5-yv77 · Cited by 1
Yuh, W. T., Ryu, S., Kim, T., & Choi, I. (2026) · Journal of Korean Neurosurgical Society · doi:10.3340/jkns.2026.0029 · Cited by 1
“Critically, this process is self-directed : experts recognize their own knowledge gaps, actively seek new information, and refine their skills through exploration without external instruction 82) .”
Candia‐Rivera, D. (2026) · Physics of Life Reviews · doi:10.1016/j.plrev.2026.06.003 · Cited by 1
“These measures capture the regulatory role of enactive mechanisms in constraining exploration while maintaining learning efficiency, consistent with work on intrinsically motivated exploration and constrained reinforcement learning [91,117].”
Wang, G., & Nixon, M. S. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-8318-8_5 · Cited by 1
Buehler, M. J. (2026) · AI for Science · doi:10.1088/3050-287x/aeb27b · Cited by 1
Proceedings of the 1st Workshop on Rights and Duties of Autonomous Agents (RDA2) 2012
Boissier, O. (2012) · Cited by 1
Bladder Endometriosis: A Case Report
Chan, C., Goykhman, I., & Horrow, M. M. (2016) · Ultrasound Quarterly · doi:10.1097/ruq.0000000000000263 · Cited by 1
Gromov, M. (2017) · Mathematics, culture, and the arts · doi:10.1007/978-3-319-53385-8_9 · Cited by 1
“…and/or curiosity driven learning: Formal Theory of Fun and Intrinsic Motivation and Creativity (1990–2010) by Jürgen Schmidhuber, http://www.idsia.ch/∼juergen/, www.ece.uvic.ca/∼bctill/papers/ememcog/Schmidhuber−2010.pdf, http://www.idsia.ch/∼juergen/, http://www.idsia.ch/∼juergen/interest.html.”
BackgroundA Hierarchical System for Autonomous Musical Creation
Reimer, M., & Garnett, G. E. (2014) · Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment · doi:10.1609/aiide.v10i5.12769 · Cited by 1
“This is derived from Schmidhuber’s theory that postulates that an improvement in a system’s ability to compress its world model is an indication of its learning or assimilating a new concept and in turn serves as an impetus for further exploration (Schmidhuber 2010).”
BackgroundInfluentialNew Tab Page Recommendations Strongly Concentrate Web Browsing to Familiar Sources
Bharadhwaj, H., & Srivastava, N. (2019) · ACM Web Science Conference (WebSci) · doi:10.1145/3292522.3326011 · Cited by 1
“more scope for diversity of experience and learning [29, 33].”
BackgroundArtificial Curiosity Driven Robots with Spatiotemporal Regularity Discovery Ability
Kalhor, D., & Loo, C. K. (2014) · Lecture notes in computer science · doi:10.1007/978-3-319-09339-0_9 · Cited by 1
Guided Reinforcement Learning via Sequence Learning
Ramamurthy, R., Sifa, R., Lübbering, M., & Bauckhage, C. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-61616-8_27 · Cited by 1
Domain Curiosity: Learning Efficient Data Collection Strategies for Domain Adaptation
Arndt, K., Struckmeier, O., & Kyrki, V. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9635864 · Cited by 1
“, learnable) uncertainty, this method was later modified to reward model improvement [15], [16], model parameter updates [17], work in feature space [18] and reward disagreement between ensemble dynamics models [19], [20],”
MethodMühlenhoff, M., & Rudloff, D. (2021) · Springer eBooks · doi:10.1007/978-3-658-28959-1_21 · Cited by 1
REAL 2021 – Robot open-Ended Autonomous Learning: A Competition and Benchmark
Baldassarre, G., Montella, D., Santucci, V. G., & Cartoni, E. (2021) · IEEE International Conference on Development and Learning (ICDL) · doi:10.1109/icdl49984.2021.9515624 · Cited by 1
Perspectives on human natural intelligence in light of trends in transhumanism
Maslova, A. V. (2022) · Research Result Social Studies and Humanities · doi:10.18413/2408-932x-2022-8-2-0-11 · Cited by 1
Spatial Consciousness Model of Intrinsic Reward in Partially Observable Environments
Ni, Z., Jin, Y., Liu, P., & Zhao, W. (2022) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-022-01771-5 · Cited by 1
“the observation with insufficient knowledge to encourage the agent to explore it, and obtain better navigation strategies after fully exploring the environment[11, 12].”
BackgroundCuriously exploring affordance spaces of a pouring task
Pomarlan, M., Hedblom, M. M., & Porzel, R. (2022) · Expert Systems · doi:10.1111/exsy.13213 · Cited by 2
“One account of intrinsic motivation is the work of Schmidhuber (Schmidhuber, 2010), in which prediction accuracy is enhanced based on the principle of maximising intrinsic rewards.”
BackgroundA Derived Functional Framework for Cognitive Computational Learning System
Feng, J., Zhang, & Luo (2026) · Cognitive Computation · doi:10.1007/s12559-026-10575-w · Cited by 1
Towards Efficient Knowledge Reuse for Open-ended Learning in Real Robots through Motivation
Romero, A., Bellas, F., Becerra, J. A., & Duro, R. J. (2023) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn54540.2023.10191570 · Cited by 1
“In particular, it has been used, among other applications, to enhance state-space exploration [10], drive systems towards knowledge gathering [11] and autonomous skill learning [12], or bias autonomous goal selection [13][14].”
MethodMnemonic Dictionary Learning for Intrinsic Motivation in Reinforcement Learning
Yan, R., Wu, Z., Zhan, Y., Tao, P., Wang, Z., Cai, Y., & Xing, J. (2023) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn54540.2023.10191424 · Cited by 2
“These methods contribute to forming a class of intrinsic motivation algorithms for reinforcement learning [12–18].”
MethodScaling Goal-based Exploration via Pruning Proto-goals
Bagaria, A., & Schaul, T. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/384 · Cited by 4
Lo, K., & DeLong, M. (2025) · Fashion Practice · doi:10.1080/17569370.2024.2433114 · Cited by 1
“The idea that intrinsic motivation fosters creativity has been substantiated by numerous studies over the past few decades (Auger and Woodman 2016; Kasof et al. 2007; Schmidhuber 2010).”
Contrastive Retrospection: honing in on critical steps for rapid learning and generalization in RL
Sun, C., Yang, W., Jiralerspong, T., Malenfant, D., Alsbury-Nealy, B., Bengio, Y., & Richards, B. (2023) · neural information processing systems · doi:10.52202/075280-1356 · Cited by 6
“Here, we chose the reward shaping approach (52; 47; 42), wherein we add additional intrinsic rewards ˜ r kt to the rewards from the environment, r kt , though other strategies are possible.”
BackgroundMethodPanta Rhei: Curiosity-Driven Exploration to Learn the Image-Schematic Affordances of Pouring Liquids
Pomarlan, M., Hedblom, M. M., & Porzel, R. (2021) · Media (https://www.suub.uni-bremen.de/) · doi:10.26092/elib/4195 · Cited by 1
A mathematical theory of knowledge, science, bias and pseudosciencePreprint
Fanelli, D. (2016) · doi:10.7287/peerj.preprints.1968v1 · Cited by 1
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.