Probabilistic photonic computing for AI DOI
Frank Brückerhoff‐Plückelmann, Anna P. Ovvyan, Akhil Varri

et al.

Nature Computational Science, Journal Year: 2025, Volume and Issue: unknown

Published: May 23, 2025

Language: Английский

Probabilistic photonic computing with chaotic light DOI Creative Commons
Frank Brückerhoff‐Plückelmann, Hendrik Borras, Bernhard Klein

et al.

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Dec. 1, 2024

Abstract Biological neural networks effortlessly tackle complex computational problems and excel at predicting outcomes from noisy, incomplete data. Artificial (ANNs), inspired by these biological counterparts, have emerged as powerful tools for deciphering intricate data patterns making predictions. However, conventional ANNs can be viewed “point estimates” that do not capture the uncertainty of prediction, which is an inherently probabilistic process. In contrast, treating ANN a model derived via Bayesian inference poses significant challenges deterministic computing architectures. Here, we use chaotic light in combination with incoherent photonic processing to enable high-speed computation quantification. We exploit architecture simultaneously perform image classification prediction network. Our prototype demonstrates seamless cointegration physical entropy source enables ultrafast parallel sampling.

Language: Английский

Citations

3

Stochastic logic in biased coupled photonic probabilistic bits DOI Creative Commons
Michael Horodynski, Charles Roques‐Carmes, Yannick Salamin

et al.

Communications Physics, Journal Year: 2025, Volume and Issue: 8(1)

Published: Jan. 20, 2025

Abstract Optical computing often employs tailor-made hardware to implement specific algorithms, trading generality for improved performance in key aspects like speed and power efficiency. An important approach that is still missing its corresponding optical probabilistic computing, used e.g. solving difficult combinatorial optimization problems. In this study, we propose an experimentally viable photonic solve arbitrary Our method relies on the insight coherent Ising machines composed of coupled biased parametric oscillators can emulate stochastic logic. We demonstrate feasibility our by using numerical simulations equivalent full density matrix formulation oscillators.

Language: Английский

Citations

0

Probabilistic photonic computing for AI DOI
Frank Brückerhoff‐Plückelmann, Anna P. Ovvyan, Akhil Varri

et al.

Nature Computational Science, Journal Year: 2025, Volume and Issue: unknown

Published: May 23, 2025

Language: Английский

Citations

0