Photonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patterns
Abstract Disorder is a pervasive characteristic of natural systems, offering a wealth of non-repeating patterns. In this study, we present a novel storage method that harnesses naturally-occurring random structures to store an arbitrary pattern in a memory device. This method, the Stochastic Emergen...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-023-44498-z |
| _version_ | 1827382071447781376 |
|---|---|
| author | Marco Leonetti Giorgio Gosti Giancarlo Ruocco |
| author_facet | Marco Leonetti Giorgio Gosti Giancarlo Ruocco |
| author_sort | Marco Leonetti |
| collection | DOAJ |
| description | Abstract Disorder is a pervasive characteristic of natural systems, offering a wealth of non-repeating patterns. In this study, we present a novel storage method that harnesses naturally-occurring random structures to store an arbitrary pattern in a memory device. This method, the Stochastic Emergent Storage (SES), builds upon the concept of emergent archetypes, where a training set of imperfect examples (prototypes) is employed to instantiate an archetype in a Hopfield-like network through emergent processes. We demonstrate this non-Hebbian paradigm in the photonic domain by utilizing random transmission matrices, which govern light scattering in a white-paint turbid medium, as prototypes. Through the implementation of programmable hardware, we successfully realize and experimentally validate the capability to store an arbitrary archetype and perform classification at the speed of light. Leveraging the vast number of modes excited by mesoscopic diffusion, our approach enables the simultaneous storage of thousands of memories without requiring any additional fabrication efforts. Similar to a content addressable memory, all stored memories can be collectively assessed against a given pattern to identify the matching element. Furthermore, by organizing memories spatially into distinct classes, they become features within a higher-level categorical (deeper) optical classification layer. |
| first_indexed | 2024-03-08T14:14:12Z |
| format | Article |
| id | doaj.art-72a9066300444099a2a8b7a7e4524cba |
| institution | Directory Open Access Journal |
| issn | 2041-1723 |
| language | English |
| last_indexed | 2024-03-08T14:14:12Z |
| publishDate | 2024-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj.art-72a9066300444099a2a8b7a7e4524cba2024-01-14T12:29:43ZengNature PortfolioNature Communications2041-17232024-01-0115111010.1038/s41467-023-44498-zPhotonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patternsMarco Leonetti0Giorgio Gosti1Giancarlo Ruocco2Soft and Living Matter Laboratory, Institute of NanotechnologySoft and Living Matter Laboratory, Institute of NanotechnologyCenter for Life Nano- & Neuro-Science, Italian Institute of TechnologyAbstract Disorder is a pervasive characteristic of natural systems, offering a wealth of non-repeating patterns. In this study, we present a novel storage method that harnesses naturally-occurring random structures to store an arbitrary pattern in a memory device. This method, the Stochastic Emergent Storage (SES), builds upon the concept of emergent archetypes, where a training set of imperfect examples (prototypes) is employed to instantiate an archetype in a Hopfield-like network through emergent processes. We demonstrate this non-Hebbian paradigm in the photonic domain by utilizing random transmission matrices, which govern light scattering in a white-paint turbid medium, as prototypes. Through the implementation of programmable hardware, we successfully realize and experimentally validate the capability to store an arbitrary archetype and perform classification at the speed of light. Leveraging the vast number of modes excited by mesoscopic diffusion, our approach enables the simultaneous storage of thousands of memories without requiring any additional fabrication efforts. Similar to a content addressable memory, all stored memories can be collectively assessed against a given pattern to identify the matching element. Furthermore, by organizing memories spatially into distinct classes, they become features within a higher-level categorical (deeper) optical classification layer.https://doi.org/10.1038/s41467-023-44498-z |
| spellingShingle | Marco Leonetti Giorgio Gosti Giancarlo Ruocco Photonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patterns Nature Communications |
| title | Photonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patterns |
| title_full | Photonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patterns |
| title_fullStr | Photonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patterns |
| title_full_unstemmed | Photonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patterns |
| title_short | Photonic Stochastic Emergent Storage for deep classification by scattering-intrinsic patterns |
| title_sort | photonic stochastic emergent storage for deep classification by scattering intrinsic patterns |
| url | https://doi.org/10.1038/s41467-023-44498-z |
| work_keys_str_mv | AT marcoleonetti photonicstochasticemergentstoragefordeepclassificationbyscatteringintrinsicpatterns AT giorgiogosti photonicstochasticemergentstoragefordeepclassificationbyscatteringintrinsicpatterns AT giancarloruocco photonicstochasticemergentstoragefordeepclassificationbyscatteringintrinsicpatterns |