paper:doi-10-1073-pnas-2000807117Supervised learning through physical changes in a mechanical system
Original abstract (expand)
Mechanical metamaterials are usually designed to show desired responses to prescribed forces. In some applications, the desired force-response relationship is hard to specify exactly, but examples of forces and desired responses are easily available. Here, we propose a framework for supervised learning in thin, creased sheets that learn the desired force-response behavior by physically experiencing training examples and then, crucially, respond correctly (generalize) to previously unseen test forces. During training, we fold the sheet using training forces, prompting local crease stiffnesses to change in proportion to their experienced strain. We find that this learning process reshapes nonlinearities inherent in folding a sheet so as to show the correct response for previously unseen test forces. We show the relationship between training error, test error, and sheet size (model complexity) in learning sheets and compare them to counterparts in machine-learning algorithms. Our framework shows how the rugged energy landscape of disordered mechanical materials can be sculpted to show desired force-response behaviors by a local physical learning process.
Related work— refs + corpus + external arXiv
Cited / in-corpus / arXiv badges show which signals surfaced each row. Multi-source rows weighted higher.
- Learning without neurons in physical systemsin corpus2022≈ 74%
- ≈ 72%
- A Comprehensive Exploration of Personalized Learning in Smart Education: From Student Modeling to Personalized RecommendationsYang Cao, Runze Li, Jiajun Cui, Hong Qian, Bo Jiang, Wei Zhang Siyu Wu2026≈ 70%
- AI, Meet Human: Learning Paradigms for Hybrid Decision Making SystemsRoberto Pellungrini, Mattia Setzu, Fosca Giannotti and Dino Pedreschi Clara Punzi2026≈ 70%
- The principles of adaptation in organisms and machines I: machine learning, information theory, and thermodynamicsHideaki Shimazaki2019≈ 70%
- Modeling motor control in continuous-time Active Inference: a surveyFederico Maggiore, Antonella Maselli, Francesco Donnarumma, Domenico Maisto, Francesco Mannella, Ivilin Peev Stoianov and Giovanni Pezzulo Matteo Priorelli2024≈ 70%
- ≈ 70%
- A Pattern Language for Machine Learning TasksIan Fan, Tuomas Laakkonen, Neil John Ortega, Thomas Hoffmann, Vincent Wang-Mascianica Benjamin Rodatz2025≈ 69%
- Robot Learning from a Physical World ModelSicheng He, Hao-Ning Wu, Yang You, Shuyang Sun, Zhicheng Wang, Yanan Bao, Huizhong Chen, Leonidas Guibas, Vitor Guizilini, Howard Zhou, and Yue Wang Jiageng Mao2025≈ 69%
- Informed POMDP: Leveraging Additional Information in Model-Based RLAdrien Bolland, Damien Ernst Gaspard Lambrechts2025≈ 69%
- What Neuroscience Can Teach AI About Learning in Continuously Changing EnvironmentsBruno Averbeck, Georgia Koppe Daniel Durstewitz2025≈ 69%
- Microelectronic Morphogenesis: Progress towards Artificial OrganismsDaniil Karnaushenko, Minshen Zhu and Oliver G. Schmidt John S. McCaskill2024≈ 69%
- Discovering Latent States for Model Learning: Applying Sensorimotor Contingencies Theory and Predictive Processing to Model ContextNikolas J. Hemion2016≈ 68%
- ≈ 68%
- There Will Be a Scientific Theory of Deep LearningDaniel Kunin, Alexander Atanasov, Enric Boix-Adser\`a, Blake Bordelon, Jeremy Cohen, Nikhil Ghosh, Florentin Guth, Arthur Jacot, Mason Kamb, Dhruva Karkada, Eric J. Michaud, Berkan Ottlik, Joseph Turnbull Jamie Simon2026≈ 68%
- Predicting Brain Morphogenesis via Physics-Transfer LearningYingjie Zhao and Yicheng Song and Fan Xu and Zhiping Xu2025≈ 68%
- ≈ 68%
- ≈ 66%
- ≈ 65%
- ≈ 65%
- Why Learning Requires Feelingin corpus2026≈ 65%
- Model Alignment Searchin corpus2025≈ 65%
- Collective intelligence: A unifying concept for integrating biology across scales and substratesin corpus2024≈ 65%
- Cognitive glues are shared models of relative scarcities: the economics of collective intelligencein corpus2026≈ 64%
- ≈ 64%
- ≈ 64%
- The World Inside Neural Networksin corpus2026≈ 64%
- ≈ 64%
- ≈ 64%
Similar preprints — Semantic Scholar
Cited by (3)
- Developmental Bioelectricity: the cognitive glue enabling evolutionary scaling from physiology to mind
Developmental bioelectricity—the network of ion channels, gap junctions, and neurotransmitter-mediated Vmem (membrane potential) dynamics operating across all body cells, not just neurons—functions as
- Learning without neurons in physical systems
Physical systems — elastic networks, flow networks, molecular assemblies, and creased sheets — can autonomously learn desired input-output behaviors through purely local learning rules, without any co
- Brains and where else? Mapping theories of consciousness to unconventional embodiments
Rouleau and Levin's 2026 analysis in *Philosophical Transactions of the Royal Society A* (384, issue 2320) demonstrates that the functional and operational principles underlying most contemporary theo