{"id":100701,"date":"2020-01-11T09:26:58","date_gmt":"2020-01-11T17:26:58","guid":{"rendered":"https:\/\/lifeboat.com\/blog\/2020\/01\/wave-physics-as-an-analog-recurrent-neural-network"},"modified":"2020-01-14T06:50:36","modified_gmt":"2020-01-14T14:50:36","slug":"wave-physics-as-an-analog-recurrent-neural-network","status":"publish","type":"post","link":"https:\/\/lifeboat.com\/blog\/2020\/01\/wave-physics-as-an-analog-recurrent-neural-network","title":{"rendered":"Wave physics as an analog recurrent neural network"},"content":{"rendered":"

<\/a><\/p>\n

Analog machine learning hardware offers a promising alternative<\/a> to digital counterparts as a more energy efficient and faster platform. Wave physics based on acoustics and optics<\/a> is a natural candidate to build analog processors for time-varying signals. In a new report on Science<\/i> Advances <\/i>Tyler W. Hughes and a research team in the departments of Applied Physics and Electrical Engineering at Stanford University, California, identified mapping between the dynamics of wave physics and computation in recurrent neural networks.<\/p>\n

The map indicated the possibility of training physical wave systems to learn complex features in temporal data using standard training techniques used for neural networks<\/a>. As proof of principle, they demonstrated an inverse-designed, inhomogeneous medium to perform English vowel classification based on raw audio signals as their waveforms scattered and propagated through it. The scientists achieved performance comparable to a standard digital implementation of a recurrent neural network. The findings will pave the way for a new class of analog machine learning<\/a> platforms for fast and efficient information processing within its native domain.<\/p>\n

The recurrent neural network<\/a> (RNN) is an important machine learning model<\/a> widely used to perform tasks including natural language processing<\/a> and time series prediction<\/a>. The team trained wave-based physical systems to function as an RNN and passively process signals and information in their native domain without analog-to-digital conversion<\/a>. The work resulted in a substantial gain in speed and reduced power consumption. In the present framework, instead of implementing circuits to deliberately route signals back to the input, the recurrence relationship<\/a> occurred naturally in the time dynamics of the physics itself. The device provided the memory capacity for information processing based on the waves as they propagated through space.<\/p>\n","protected":false},"excerpt":{"rendered":"

Analog machine learning hardware offers a promising alternative to digital counterparts as a more energy efficient and faster platform. Wave physics based on acoustics and optics is a natural candidate to build analog processors for time-varying signals. In a new report on Science Advances Tyler W. Hughes and a research team in the departments of [\u2026]<\/p>\n","protected":false},"author":511,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[38,1965,219,6],"tags":[],"class_list":["post-100701","post","type-post","status-publish","format-standard","hentry","category-engineering","category-mapping","category-physics","category-robotics-ai"],"_links":{"self":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/100701","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/users\/511"}],"replies":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/comments?post=100701"}],"version-history":[{"count":1,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/100701\/revisions"}],"predecessor-version":[{"id":100839,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/100701\/revisions\/100839"}],"wp:attachment":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/media?parent=100701"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/categories?post=100701"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/tags?post=100701"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}