Limits of nonlinear and dispersive fiber propagation for photonic extreme learning

2025-03-05

AI Summary

This work investigates the fundamental performance limits of photonic extreme learning machines (ELMs) operating in nonlinear dispersive optical fiber. To this end, a high-fidelity fiber propagation model is developed based on the nonlinear Schrödinger equation, enabling systematic evaluation of propagation dynamics, spectral encoding schemes, readout mechanisms, and noise—particularly quantum noise—on classification accuracy. The study reveals, for the first time, a significant performance disparity between anomalous and normal dispersion regimes: handwritten digit classification accuracies reach 91% and 93%, respectively, with the normal dispersion regime demonstrating superior performance. Moreover, it quantitatively establishes quantum noise as an intrinsic source of irreducible accuracy degradation, imposing a fundamental precision ceiling. By integrating rigorous nonlinear optical modeling with machine learning benchmarking, this work provides both theoretical foundations and experimental benchmarks for assessing the physical realizability and noise robustness of photonic neural networks.

Technology Category

• Machine Learning: Quantum Machine Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

We report a generalized nonlinear Schr""odinger equation simulation model of an extreme learning machine based on optical fiber propagation. Using handwritten digit classification as a benchmark, we study how accuracy depends on propagation dynamics, as well as parameters governing spectral encoding, readout, and noise. Test accuracies of over 91% and 93% are found for propagation in the anomalous and normal dispersion regimes respectively. Our simulation results also suggest that quantum noise on the input pulses introduces an intrinsic penalty to ELM performance.

Problem

• Modeling optical fiber-based extreme learning machine dynamics
• Analyzing accuracy dependence on propagation and encoding parameters
• Investigating quantum noise impact on machine learning performance

Innovation

• Generalized nonlinear Schrödinger equation simulation model
• Optical fiber propagation for extreme learning
• Handwritten digit classification benchmark testing