Low-power scalable multilayer optoelectronic neural networks enabled with incoherent light
Alexander Song, Sai Nikhilesh Murty Kottapalli, Rahul Goyal, Bernhard Schölkopf, Peer Fischer · 2024
Read Low-power scalable multilayer optoelectronic neural networks enabled with incoherent light by Alexander Song, Sai Nikhilesh Murty Kottapalli, Rahul Goyal, Bernhard Schölkopf, Peer Fischer (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11655539.txt.
Description
Alexander Song and a team of co-authors build a neural network that computes with light and electronics together, in two thousand twenty-four. Optical computing promises speed and low energy for modern learning tasks. Reading data in and out, though, has held back earlier optical designs.
Their framework swaps between optical layers and optoelectronic layers to carry out the core math steps. Flat arrays of tiny lights and detectors, linked by simple analog circuits, run the layers in parallel and in real time. A three-layer build recognized handwritten digits with high accuracy and sorted a curved data set well. Doing many layers at once cuts the number of read-ins and read-outs sharply.
Catalog record
| CREATOR | Alexander Song, Sai Nikhilesh Murty Kottapalli, Rahul Goyal, Bernhard Schölkopf, Peer Fischer |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |