Artificial Neural Networks for Photonic Applications: From Algorithms to Implementation

Abstract

This tutorial-review on applications of artificial neural networks in photonics targets a broad audience, ranging from optical research and engineering communities to computer science and applied mathematics. We focus here on the research areas at the interface between these disciplines, attempting to find the right balance between technical details specific to each domain and overall clarity. First, we briefly recall key properties and peculiarities of some core neural network types, which we believe are the most relevant to photonics, also linking the layer's theoretical design to some photonics hardware realizations. After that, we elucidate the question of how to fine-tune the selected model's design to perform the required task with optimized accuracy. Then, in the review part, we discuss recent developments and progress for several selected applications of neural networks in photonics, including multiple aspects relevant to optical communications, imaging, sensing, and the design of new materials and lasers. In the following section, we put a special emphasis on how to accurately evaluate the complexity of neural networks in the context of the transition from algorithms to hardware implementation. The introduced complexity characteristics are used to analyze the applications of neural networks in optical communications, as a specific, albeit highly important example, comparing those with some benchmark signal processing methods. We combine the description of the well-known model compression strategies used in machine learning, with some novel techniques introduced recently in optical applications of neural networks. It is important to stress that although our focus in this tutorial-review is on photonics, we believe that the methods and techniques presented here can be handy in a much wider range of scientific and engineering applications.

References in corpus (85)

• Distilling the Knowledge in a Neural Network
• Knowledge Distillation: A Survey
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
• Deep Learning with Coherent Nanophotonic Circuits
• End to End Learning for Self-Driving Cars
• A Critical Review of Recurrent Neural Networks for Sequence Learning
• Outlook for inverse design in nanophotonics
• Photonics for artificial intelligence and neuromorphic computing
• Tutorial on Variational Autoencoders
• Rethinking the Value of Network Pruning
• Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users
• A Survey of Model Compression and Acceleration for Deep Neural Networks
• Neuromorphic Silicon Photonic Networks
• A Review of Convolutional Neural Networks for Inverse Problems in Imaging
• Brain-inspired computing: We need a master plan
• Deep neural networks for the evaluation and design of photonic devices
• Highway Networks
• Deep Reinforcement Learning: An Overview
• Training of photonic neural networks through in situ backpropagation
• Deep learning in nano-photonics: inverse design and beyond
• Global optimization of dielectric metasurfaces using a physics-driven neural network
• Fast inference of deep neural networks in FPGAs for particle physics
• End-to-end Deep Learning of Optical Fiber Communications
• Reinforcement Learning in a large scale photonic Recurrent Neural Network
• Wave Physics as an Analog Recurrent Neural Network
• Experimentally realized in situ backpropagation for deep learning in nanophotonic neural networks
• Hardware-oriented Approximation of Convolutional Neural Networks
• An Overview of Deep Semi-Supervised Learning
• Information Transmission using the Nonlinear Fourier Transform, Part I: Mathematical Tools
• Machine-Learning-Assisted Metasurface Design for High-Efficiency Thermal Emitter Optimization
• Transmission of natural scene images through a multimode fibre
• Realistic Evaluation of Deep Semi-Supervised Learning Algorithms
• Mechanisms of Spatiotemporal Mode-Locking
• Photonic tensor cores for machine learning
• Applying Neural Networks in Optical Communication Systems: Possible Pitfalls
• Learning to decompose the modes in few-mode fibers with deep convolutional neural network
• Machine Learning for Intelligent Optical Networks: A Comprehensive Survey
• Deep Learning and Model Predictive Control for Self-Tuning Mode-Locked Lasers
• Towards 'smart lasers': self-optimisation of an ultrafast pulse source using a genetic algorithm
• Turbulence correction with artificial neural networks
• Nonlinear inverse synthesis for high spectral efficiency transmission in optical fibers
• Compensation of Fiber Nonlinearities in Digital Coherent Systems Leveraging Long Short-Term Memory Neural Networks
• Autoencoders
• Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networks
• Fast and Accurate Optical Fiber Channel Modeling Using Generative Adversarial Network
• UNIQ: Uniform Noise Injection for Non-Uniform Quantization of Neural Networks
• End-to-End Optimized Transmission over Dispersive Intensity-Modulated Channels Using Bidirectional Recurrent Neural Networks
• Performance and Complexity Analysis of bi-directional Recurrent Neural Network Models vs. Volterra Nonlinear Equalizers in Digital Coherent Systems
• Scientific Machine Learning through Physics-Informed Neural Networks: Where we are and What's next
• Ultralow complexity long short-term memory network for fiber nonlinearity mitigation in coherent optical communication systems
• Optical Fringe Patterns Filtering Based on Multi-Stage Convolution Neural Network
• 400 Gbps Dual-polarisation Non-linear Frequency-division Multiplexed Transmission with b-Modulation
• Integrated Photonic Tensor Processing Unit for a Matrix Multiply: a Review
• Genetic algorithm-based control of birefringent filtering for self-tuning, self-pulsing fiber lasers
• SIMBA: Scalable Inversion in Optical Tomography using Deep Denoising Priors
• End-to-end optimization of coherent optical communications over the split-step Fourier method guided by the nonlinear Fourier transform theory
• Multimode Mamyshev Oscillator
• Deep Learning Models for Global Coordinate Transformations that Linearize PDEs
• Principle-driven Fiber Transmission Model based on PINN Neural Network
• A Review of Designs and Applications of Echo State Networks
• A Survey of Quantization Methods for Efficient Neural Network Inference
• Computational Complexity Evaluation of Neural Network Applications in Signal Processing
• Machine Learning Framework for Quantum Sampling of Highly-Constrained, Continuous Optimization Problems
• Enhanced Atmospheric Turbulence Resiliency with Successive Interference Cancellation DSP in Mode Division Multiplexing Free-Space Optical Links
• ShiftAddNet: A Hardware-Inspired Deep Network
• Understanding Domain Randomization for Sim-to-real Transfer
• Gradient-free training of autoencoders for non-differentiable communication channels
• Knowledge Distillation Circumvents Nonlinearity for Optical Convolutional Neural Networks
• Wiener-Hammerstein model and its learning for nonlinear digital pre-distortion of optical transmitters
• Neural Network Quantization for Efficient Inference: A Survey
• Exploiting the ConvLSTM: Human Action Recognition using Raw Depth Video-Based Recurrent Neural Networks
• Towards Efficient Post-training Quantization of Pre-trained Language Models
• On the Statistical Challenges of Echo State Networks and Some Potential Remedies
• Dispersion Characterization and Pulse Prediction with Machine Learning
• Spectral and spatial power evolution design with machine learning-enabled Raman amplification
• Recurrent Neural Processes
• Forecasting Loss of Signal in Optical Networks with Machine Learning
• High-Cardinality Hybrid Shaping for 4D Modulation Formats in Optical Communications Optimized via End-to-End Learning
• Methods for Pruning Deep Neural Networks
• Nonlinear Equalization for Optical Communications Based on Entropy-Regularized Mean Square Error
• Application of Transformers for Nonlinear Channel Compensation in Optical Systems
• Deep learning for enhanced free-space optical communications
• Spiking Neural Network Equalization for IM/DD Optical Communication
• Knowledge Distillation Applied to Optical Channel Equalization: Solving the Parallelization Problem of Recurrent Connection

Cited by in corpus (2)

• Conditional Recurrent Neural Networks for broad applications in nonlinear optics
• Optimized higher-order photon state classification by machine learning