Fault tolerance and noise immunity in freespace diffractive optical neural networks

Show simple item record

dc.contributor.author Panda, Soumyashree S.
dc.contributor.author Hegde, Ravi S.
dc.coverage.spatial United Kingdom
dc.date.accessioned 2022-04-19T06:30:50Z
dc.date.available 2022-04-19T06:30:50Z
dc.date.issued 2022-03
dc.identifier.citation Panda, Soumyashree S. and Hegde, Ravi S., "Fault tolerance and noise immunity in freespace diffractive optical neural networks", Engineering Research Express, DOI: 10.1088/2631-8695/ac4832, vol. 4, no. 1, Mar. 2022. en_US
dc.identifier.issn 2631-8695
dc.identifier.uri https://doi.org/10.1088/2631-8695/ac4832
dc.identifier.uri https://repository.iitgn.ac.in/handle/123456789/7658
dc.description.abstract Free-space diffractive optical networks are a class of trainable optical media that are currently being explored as a novel hardware platform for neural engines. The training phase of such systems is usually performed in a computer and the learned weights are then transferred onto optical hardware ('ex-situ training'). Although this process of weight transfer has many practical advantages, it is often accompanied by performance degrading faults in the fabricated hardware. Being analog systems, these engines are also subject to performance degradation due to noises in the inputs and during optoelectronic conversion. Considering diffractive optical networks trained for image classification tasks on standard datasets, we numerically study the performance degradation arising out of weight faults and injected noises and methods to ameliorate these effects. Training regimens based on intentional fault and noise injection during the training phase are only found marginally successful at imparting fault tolerance or noise immunity. We propose an alternative training regimen using gradient based regularization terms in the training objective that are found to impart some degree of fault tolerance and noise immunity in comparison to injection based training regimen.
dc.description.statementofresponsibility by Soumyashree S. Panda and Ravi S. Hegde
dc.format.extent vol. 4, no. 1
dc.language.iso en_US en_US
dc.publisher IOP Publishing en_US
dc.subject All optical neural network en_US
dc.subject Neural network hardware en_US
dc.subject Engineered photonic materials en_US
dc.subject Inverse design en_US
dc.subject Trainable photonic media en_US
dc.subject Freespace diffractive optics en_US
dc.title Fault tolerance and noise immunity in freespace diffractive optical neural networks en_US
dc.type Article en_US
dc.relation.journal Engineering Research Express


Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search Digital Repository


Browse

My Account