{"title":"Multistable Physical Neural Networks","authors":"Eran Ben-Haim, Sefi Givli, Yizhar Or, Amir Gat","doi":"arxiv-2406.00082","DOIUrl":null,"url":null,"abstract":"Artificial neural networks (ANNs), which are inspired by the brain, are a\ncentral pillar in the ongoing breakthrough in artificial intelligence. In\nrecent years, researchers have examined mechanical implementations of ANNs,\ndenoted as Physical Neural Networks (PNNs). PNNs offer the opportunity to view\ncommon materials and physical phenomena as networks, and to associate\ncomputational power with them. In this work, we incorporated mechanical\nbistability into PNNs, enabling memory and a direct link between computation\nand physical action. To achieve this, we consider an interconnected network of\nbistable liquid-filled chambers. We first map all possible equilibrium\nconfigurations or steady states, and then examine their stability. Building on\nthese maps, both global and local algorithms for training multistable PNNs are\nimplemented. These algorithms enable us to systematically examine the network's\ncapability to achieve stable output states and thus the network's ability to\nperform computational tasks. By incorporating PNNs and multistability, we can\ndesign structures that mechanically perform tasks typically associated with\nelectronic neural networks, while directly obtaining physical actuation. The\ninsights gained from our study pave the way for the implementation of\nintelligent structures in smart tech, metamaterials, medical devices, soft\nrobotics, and other fields.","PeriodicalId":501305,"journal":{"name":"arXiv - PHYS - Adaptation and Self-Organizing Systems","volume":"35 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Adaptation and Self-Organizing Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2406.00082","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Artificial neural networks (ANNs), which are inspired by the brain, are a
central pillar in the ongoing breakthrough in artificial intelligence. In
recent years, researchers have examined mechanical implementations of ANNs,
denoted as Physical Neural Networks (PNNs). PNNs offer the opportunity to view
common materials and physical phenomena as networks, and to associate
computational power with them. In this work, we incorporated mechanical
bistability into PNNs, enabling memory and a direct link between computation
and physical action. To achieve this, we consider an interconnected network of
bistable liquid-filled chambers. We first map all possible equilibrium
configurations or steady states, and then examine their stability. Building on
these maps, both global and local algorithms for training multistable PNNs are
implemented. These algorithms enable us to systematically examine the network's
capability to achieve stable output states and thus the network's ability to
perform computational tasks. By incorporating PNNs and multistability, we can
design structures that mechanically perform tasks typically associated with
electronic neural networks, while directly obtaining physical actuation. The
insights gained from our study pave the way for the implementation of
intelligent structures in smart tech, metamaterials, medical devices, soft
robotics, and other fields.