English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Plastic self-adaptation, nonlinear recurrent dynamics and multi-scale memoryare desired features in hardware implementations of neural networks, becausethey enable them to learn, adapt and process information similarly to the waybiological brains do. In this work, we experimentally demonstrate theseproperties occurring in arrays of photonic neurons. Importantly, this isrealised autonomously in an emergent fashion, without the need for an externalcontroller setting weights and without explicit feedback of a global rewardsignal. Using a hierarchy of such arrays coupled to a backpropagation-freetraining algorithm based on simple logistic regression, we are able to achievea performance of 98.2% on the MNIST task, a popular benchmark task looking atclassification of written digits. The plastic nodes consist of siliconphotonics microring resonators covered by a patch of phase-change material thatimplements nonvolatile memory. The system is compact, robust, andstraightforward to scale up through the use of multiple wavelengths. Moreover,it constitutes a unique platform to test and efficiently implement biologicallyplausible learning schemes at a high processing speed.

Emergent self-adaptation in an integrated photonic neural network for backpropagation-free learning