cs.LGDec 1, 2025

Temporal speckle enables probabilistic weights for uncertainty-aware photonic AI

Authors: F. Brückerhoff-Plückelmann, H. Borras, S. U. Hulyal, L. Meyer, X. Ji, J. Hu, J. Sun, B. Klein, 

Organizations: Kirchhoff-Institute for Physics, University of Heidelberg; Heidelberg, 69120, Germany. · Institute of Computer Engineering, University of Heidelberg; Heidelberg, 69120, Germany. · Institute of Physics, Swiss Federal Institute of Technology, Lausanne, CH-1015 Lausanne, Switzerland.

Abstract

Artificial intelligence increasingly guides decisions in safety-critical settings, yet neural networks can remain overconfident when confronted with ambiguous observations or unfamiliar inputs. Bayesian neural networks quantify such uncertainty by replacing every weight with a probability distribution, but sampling many distributions at prediction time overwhelms deterministic digital hardware. Here we show that the temporal speckle of broadband light provides a native physical representation of probabilistic weights. Random beating among the frequency components of individually amplitude-bandwidth programmed spectral channels enables optical-encoding of mean and variance in each channel independently. This way, a single broadband light source stores many probabilistic weights in parallel and delivers mutually uncorrelated samples without requiring computationally demanding random-number generation. We harness this mechanism in an ultrafast photonic processor that computes one probabilistic convolution every 37.5 ps and allows simultaneous classification and out-of-distribution detection in hybrid Bayesian neural networks. Our system classifies blood-cell images while recognizing withheld cell types as unfamiliar, and separates epistemic novelty from aleatoric ambiguity on standard uncertainty benchmarks. Programming probability distributions into the fluctuations of light removes the sampling bottleneck of Bayesian inference and opens a route to high-speed, uncertainty-aware photonic AI.

Explore similar work

CardsList
  1. Constrained Co-Design for Photonic Bayesian Neural Networks

    Aug 3, 2026Hendrik Borras, Xiao Wang, Bernhard Klein +4Bayesian Neural NetworksUncertainty Quantification

  2. Accelerated Execution of Bayesian Neural Networks using a Single Probabilistic Forward Pass and Code Generation

    Nov 28, 2025Bernhard Klein, Falk Selker, Hendrik Borras +3Bayesian Neural NetworksEfficient Neural Network Inference

  3. Recovering topological information of light by topological learning

    Sep 6, 2026Benquan Wang, Trishita Das, Yuhan Peng +8Topological Representation LearningComputational Imaging