Supervised learning in a mechanical system

Abstract

Mechanical metamaterials are usually designed to show desired responses to prescribed forces. In some applications, the desired force-response relationship might be hard to specify exactly, although examples of forces and corresponding desired responses are easily available. Here we propose a framework for supervised learning in a thin creased sheet that learns the desired force-response behavior from training examples of spatial force patterns and can then respond correctly to previously unseen test forces. During training, we fold the sheet using different training forces and assume a learning rule that changes stiffness of creases in response to their folding strain. We find that this learning process reshapes non-linearities inherent in folding a sheet so as to show the correct response for previously unseen test forces. We study the relationship between training error, test error and sheet size which plays the role of model complexity. Our framework shows how the complex energy landscape of disordered mechanical materials can be reshaped using an iterative local learning rule.

7 pages, 7 figures

References in corpus (13)

• An overview of gradient descent optimization algorithms
• SchNet - a deep learning architecture for molecules and materials
• Topological mechanics of gyroscopic metamaterials
• Memory formation in matter
• Multi-step self-guided pathways for shape-changing metamaterials
• Designing allostery-inspired response in mechanical networks
• Accelerated search and design of stretchable graphene kirigami using machine learning
• Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures
• Directed aging, memory and Nature's greed
• Branches of triangulated origami near the unfolded state
• Shaping dynamical folding and misfolding pathways in mechanical metamaterials
• Conformational Control of Mechanical Networks
• Design of Large Sequential Conformational Change in Mechanical Networks

Cited by in corpus (13)

• Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems
• Learning without neurons in physical systems
• Demonstration of Decentralized, Physics-Driven Learning
• Supervised learning in physical networks: From machine learning to learning machines
• Desynchronous Learning in a Physics-Driven Learning Network
• Learned multi-stability in mechanical networks
• Learning to self-fold at a bifurcation
• Physical Neural Networks with Self-Learning Capabilities
• Physical Computing for Materials Acceleration Platforms
• Physical networks become what they learn
• Kinetic theory of decentralized learning for smart active matter
• Memory in cyclically crumpled sheets
• Physical learning beyond the quasistatic limit