arXiv subjects

Atsushi Uchida

Publications and source records attributed to Atsushi Uchida.

At least 19 recordsLinked to original sources

arXiv2026

Photonic reservoir computing with complex networks

Photonic reservoir computing has attracted increasing attention as a fast and low-cost approach for time-series prediction. Photonic reservoir computing utilizes the high speed, broad bandwidth, and spatial parallelism of light. However, the effect of the internal connection structure (network topology) on the computing performance has not been investigated for large-scale photonic reservoirs. In this study, we experimentally and numerically demonstrate photonic reservoir computing using a spatial light modulator to systematically evaluate the relationship between the network topology and the performance of reservoir computing. We introduce complex network structures such as small-world and scale-free network topologies of the internal nodes in the reservoir. We perform the memory capacity measurement and the one-step-ahead prediction task of the chaotic time series to compare the performance. We found that the small-world network exhibits the maximum memory capacity and the best prediction performance. Our numerical calculations reveal that the performance of the time-series prediction can be optimized by changing the rewiring probability of the network and the leak rate of the reservoir. We also implement photonic human brain network as a reservoir, which is designed by the connectomes of human brain activities. We found that the network topology strongly affects the performance of reservoir computing, and the small-world network structure outperforms the other configurations.

physics.optics↗

arXiv2026

Approximate reservoir computing with a semiconductor laser for reducing energy consumption

Photonic reservoir computing is a promising physical machine-learning technique for predicting time-series data. The quantization of the response signal from the reservoir is required for the implementation of photonic reservoir computing, and the number of quantization bits and sampling frequency need to be optimized to achieve high performance and low energy consumption. However, few studies have been reported to investigate the effect of bit quantization and sampling frequency. In this study, we introduce a concept of approximate reservoir computing with a semiconductor laser by quantizing the amplitude of node states in the reservoir and output weights. We evaluate the performance of a chaotic time-series prediction task and energy consumption per sample. We achieve significant reduction of energy consumption by optimizing the number of quantization bits, the sampling frequency, and the injection current of the semiconductor laser, while maintaining the prediction performance.

physics.optics↗

arXiv2026

Autocorrelation effects in a stochastic-process model for solving two-armed bandit problems

Decision makers exploiting photonic chaotic dynamics obtained by semiconductor lasers provide an ultrafast approach to solving multi-armed bandit problems by using a temporal optical signal as the driving source for sequential decisions. In such systems, the sampling interval of the chaotic waveform shapes the temporal correlation of the resulting time series, and experiments have reported that decision accuracy depends strongly on this autocorrelation property. However, it remains unclear whether the benefit of autocorrelation can be explained by a minimal mathematical model. Here, we analyze a stochastic-process model for solving the two-armed bandit problem based on time series, where the threshold and a two-valued Markov signal evolve jointly. Numerical results reveal an environment-dependent structure: negative (positive) autocorrelation is optimal in reward-rich (reward-poor) environments. These findings show that negative autocorrelation of the time series is advantageous when the sum of the winning probabilities is more than one, whereas positive autocorrelation is useful when the sum of the winning probabilities is less than one. Moreover, the performance is independent of autocorrelation if the sum of the winning probabilities equals one, which is mathematically clarified. This study paves the way for solving the two-armed bandit problems for reinforcement learning applications in wireless communications and robotics.

cs.LG↗

arXiv2026

Adaptive Sensing of Continuous Physical Systems for Machine Learning

Physical dynamical systems can be viewed as natural information processors: their systems preserve, transform, and disperse input information. This perspective motivates learning not only from data generated by such systems, but also how to measure them in a way that extracts the most useful information for a given task. We propose a general computing framework for adaptive information extraction from dynamical systems, in which a trainable attention module learns both where to probe the system state and how to combine these measurements to optimize prediction performance. As a concrete instantiation, we implement this idea using a spatiotemporal field governed by a partial differential equation as the underlying dynamics, though the framework applies equally to any system whose state can be sampled. Our results show that adaptive spatial sensing significantly improves prediction accuracy on canonical chaotic benchmarks. This work provides a perspective on attention-enhanced reservoir computing as a special case of a broader paradigm: neural networks as trainable measurement devices for extracting information from physical dynamical systems.

cs.LG↗

arXiv2026

Reservoir Computing as a Language Model

Large Language Models (LLM) have dominated the science and media landscape duo to their impressive performance on processing large chunks of data and produce human-like levels of text. Nevertheless, their huge energy demand and slow processing are still a bottleneck to further increasing quality while also making the models accessible to everyone. To solve this bottleneck, we will investigate how reservoir computing performs on natural text processing, which could enable fast and energy efficient hardware implementations. Studies investigating the use of reservoir computing as a language model remain sparse. In this paper, we compare three distinct approaches for character-level language modeling, two different \emph{reservoir computing} approaches, where only an output layer is trainable, and the well-known \emph{transformer}-based architectures, which fully learn an attention-based sequence representation. We explore the performance, computational cost and prediction accuracy for both paradigms by equally varying the number of trainable parameters for all models. Using a consistent pipeline for all three approaches, we demonstrate that transformers excel in prediction quality, whereas reservoir computers remain highly efficient reducing the training and inference speed. Furthermore, we investigate two types of reservoir computing: a \emph{traditional reservoir} with a static linear readout, and an \emph{attention-enhanced reservoir} that dynamically adapts its output weights via an attention mechanism. Our findings underline how these paradigms scale and offer guidelines to balance resource constraints with performance.

cs.CL↗

arXiv2025

Scalable ultrafast random bit generation using wideband chaos-based entropy sources

The exponential growth of data transmission and processing speeds in modern digital infrastructure requires entropy sources capable of producing large volumes of true randomness for information security. Chaotic emissions from semiconductor lasers are attractive in this context because of their fast dynamics and nonrepetitive behavior. Their spectral bandwidth, however, is typically limited to several tens of gigahertz, which constrains the achievable entropy rate and makes ultrafast random bit generation difficult without substantial post-processing. Here, we demonstrate a chaos-based entropy source that employs optical heterodyning between the chaotic emission from a semiconductor laser and an optical frequency comb, yielding a bandwidth exceeding 100 GHz and an experimentally verified single-channel entropy rate of 1.86 Tb/s. By directly extracting multiple bits from the digitized output of the entropy source, we achieve a single-channel random bit generation rate of 1.536 Tb/s, while four-channel parallelization reaches 6.144 Tb/s with no observable interchannel correlation. This linear scalability suggests that aggregate throughput could reach hundreds of terabits per second with additional parallel channels. The broadband, low-overhead photonic architecture presented here provides a viable route to real-time, ultrafast random bit generation with broad implications for secure communications, high-performance AI computing, and large-scale data analytics.

physics.optics↗

arXiv2025

Expanding detection bandwidth via a photonic reservoir for ultrafast optical sensing

The detection of ultrafast optical and radio-frequency (RF) signals is crucial for applications ranging from high-speed communications to advanced sensing. However, conventional detectors are fundamentally constrained by their intrinsic bandwidth, limiting accurate broadband signal measurement. Here, we show that a neuromorphic photonic processing approach can overcome this limitation, enabling accurate broadband signal detection beyond the detector bandwidth. The key idea lies in the spatiotemporal encoding of input waveforms within a photonic reservoir network, which reconstructs high-frequency components otherwise inaccessible to individual detectors. We experimentally demonstrate the detection of high-speed optical phase signals with more than an eightfold effective bandwidth expansion using an on-chip silicon photonic reservoir. This approach provides a scalable and integrable platform for high-speed optical and RF signal processing, opening new opportunities in ultrafast photonics and next-generation communication systems.

physics.optics↗

arXiv2025

Photonic decision making using optical frequency difference detection in mutually-coupled semiconductor lasers

As electronic computing approaches its performance limits, photonic accelerators have emerged as promising alternatives. Photonic accelerators exploiting semiconductor-laser synchronization have been studied for decision-making. While conventional cross-correlation methods are computationally intensive and memory demanding, we propose a frequency-based approach using optical frequency differences. Simulations and experiments confirm that this method achieves decision making with reduced cost and memory requirements.

physics.optics↗

arXiv2025

Quantum spatial best-arm identification via quantum walks

Quantum reinforcement learning has emerged as a framework combining quantum computation with sequential decision-making, and applications to the multi-armed bandit (MAB) problem have been reported. The graph bandit problem extends the MAB setting by introducing spatial constraints, where the accessibility of arms is restricted by graph connectivity, yet quantum approaches to this setting remain limited. In this paper, we formulate best-arm identification in graph bandits and propose a quantum algorithmic framework, termed Quantum Spatial Best-Arm Identification (QSBAI), which is applicable to general graph structures. This framework uses quantum walks to encode superpositions over graph-constrained actions, thereby extending amplitude amplification and generalizing the quantum BAI algorithm via Szegedy's walk framework. We focus our theoretical analysis on complete and bipartite graphs, deriving the maximal success probability of identifying the best arm and the time step at which it is achieved. Our results clarify how quantum-walk-based search can be adapted to structurally constrained decision problems and provide a foundation for quantum best-arm identification in graph-structured environments.

quant-ph↗

arXiv2025

Attention-Enhanced Reservoir Computing as a Multiple Dynamical System Approximator

Reservoir computing has proven effective for tasks such as time-series prediction, particularly in the context of chaotic systems. However, conventional reservoir computing frameworks often face challenges in achieving high prediction accuracy and adapting to diverse dynamical problems due to their reliance on fixed weight structures. A concept of an attention-enhanced reservoir computer has been proposed, which integrates an attention mechanism into the output layer of the reservoir computing model. This addition enables the system to prioritize distinct features dynamically, enhancing adaptability and prediction performance. In this study, we demonstrate the capability of the attention-enhanced reservoir computer to learn and predict multiple chaotic attractors simultaneously with a single set of weights, thus enabling transitions between attractors without explicit retraining. The method is validated using benchmark tasks, including the Lorenz system, Rössler system, Henon map, Duffing oscillator, and Mackey-Glass delay-differential equation. Our results indicate that the attention-enhanced reservoir computer achieves superior prediction accuracy, valid prediction times, and improved representation of spectral and histogram characteristics compared to traditional reservoir computing methods, establishing it as a robust tool for modeling complex dynamical systems.

nlin.CD↗

arXiv2025

Blending Optimal Control and Biologically Plausible Learning for Noise-Robust Physical Neural Networks

The rapidly increasing computational demands for artificial intelligence (AI) have spurred the exploration of computing principles beyond conventional digital computers. Physical neural networks (PNNs) offer efficient neuromorphic information processing by harnessing the innate computational power of physical processes; however, training their weight parameters is computationally expensive. We propose a training approach for substantially reducing this training cost. Our training approach merges an optimal control method for continuous-time dynamical systems with a biologically plausible training method--direct feedback alignment. In addition to the reduction of training time, this approach achieves robust processing even under measurement errors and noise without requiring detailed system information. The effectiveness was numerically and experimentally verified in an optoelectronic delay system. Our approach significantly extends the range of physical systems practically usable as PNNs.

physics.app-ph↗

arXiv2024

Attention-Enhanced Reservoir Computing

Photonic reservoir computing has been successfully utilized in time-series prediction as the need for hardware implementations has increased. Prediction of chaotic time series remains a significant challenge, an area where the conventional reservoir computing framework encounters limitations of prediction accuracy. We introduce an attention mechanism to the reservoir computing model in the output stage. This attention layer is designed to prioritize distinct features and temporal sequences, thereby substantially enhancing the prediction accuracy. Our results show that a photonic reservoir computer enhanced with the attention mechanism exhibits improved prediction capabilities for smaller reservoirs. These advancements highlight the transformative possibilities of reservoir computing for practical applications where accurate prediction of chaotic time series is crucial.

cs.ET↗

arXiv2023

Experimental control of mode-competition dynamics in a chaotic multimode semiconductor laser for decision making

Photonic computing has been widely used to accelerate the computational performance in machine learning. Photonic decision-making is a promising approach that uses photonic computing technologies to solve the multi-armed bandit problem based on reinforcement learning. Photonic decision making using chaotic mode competition dynamics has been proposed. However, the experimental conditions for achieving superior decision-making performance have not yet been established. In this study, we experimentally investigate the mode competition dynamics in a chaotic multimode semiconductor laser in the presence of optical feedback and injection. We control chaotic mode-competition dynamics via optical injection, and we found that positive wavelength detuning results in an efficient mode concentration to one of the longitudinal modes with a small optical injection power. We experimentally investigate two-dimensional bifurcation map of the laser dynamics of the total intensity. Complex mixed dynamics are observed in the presence of optical feedback and injection. We experimentally conduct decision making to solve the bandit problem using chaotic mode-competition dynamics. A fast mode concentration property is observed at positive wavelength detuning, which resulted in a fast convergence of the correct decision rate. Our findings could be useful in accelerating the decision-making performance in adaptive optical networks using reinforcement learning.

physics.optics↗

arXiv2023

Solving multi-armed bandit problems using a chaotic microresonator comb

The Multi-Armed Bandit (MAB) problem, foundational to reinforcement learning-based decision-making, addresses the challenge of maximizing rewards amidst multiple uncertain choices. While algorithmic solutions are effective, their computational efficiency diminishes with increasing problem complexity. Photonic accelerators, leveraging temporal and spatial-temporal chaos, have emerged as promising alternatives. However, despite these advancements, current approaches either compromise computation speed or amplify system complexity. In this paper, we introduce a chaotic microresonator frequency comb (chaos comb) to tackle the MAB problem, where each comb mode is assigned to a slot machine. Through a proof-of-concept experiment, we employ 44 comb modes to address an MAB with 44 slot machines, demonstrating performance competitive with both conventional software algorithms and other photonic methods. Further, the scalability of decision making is explored with up to 512 slot machines using experimentally obtained temporal chaos in different time slots. Power-law scalability is achieved with an exponent of 0.96, outperforming conventional software-based algorithms. Moreover, we find that a numerically calculated chaos comb accurately reproduces experimental results, paving the way for discussions on strategies to increase the number of slot machines.

physics.optics↗

arXiv2023

Ultrafast single-channel machine vision based on neuro-inspired photonic computing

High-speed machine vision is increasing its importance in both scientific and technological applications. Neuro-inspired photonic computing is a promising approach to speed-up machine vision processing with ultralow latency. However, the processing rate is fundamentally limited by the low frame rate of image sensors, typically operating at tens of hertz. Here, we propose an image-sensor-free machine vision framework, which optically processes real-world visual information with only a single input channel, based on a random temporal encoding technique. This approach allows for compressive acquisitions of visual information with a single channel at gigahertz rates, outperforming conventional approaches, and enables its direct photonic processing using a photonic reservoir computer in a time domain. We experimentally demonstrate that the proposed approach is capable of high-speed image recognition and anomaly detection, and furthermore, it can be used for high-speed imaging. The proposed approach is multipurpose and can be extended for a wide range of applications, including tracking, controlling, and capturing sub-nanosecond phenomena.

physics.optics↗

arXiv2023

Analysis of temporal structure of laser chaos by Allan variance

Allan variance has been widely utilized in evaluating the stability of the time series generated by atomic clocks and lasers, in time regimes ranging from short to extremely long. This multi-scale examination capability of the Allan variance may also be beneficial in evaluating the chaotic oscillating dynamics of semiconductor lasers, not just for conventional phase stability analysis purposes. Here we demonstrate Allan variance analysis of the complex time series generated by a semiconductor laser with delayed feedback, including low-frequency fluctuations (LFFs), which exhibit both fast and slow dynamics. Whereas the detection of LFFs is not easy with the conventional power spectrum analysis method in the low-frequency regime, we show that the Allan variance approach clearly captured the appearance of multiple time-scale dynamics, like LFFs. This study demonstrates that Allan variance can help in understanding and characterizing versatile laser dynamics, including LFFs, which involve dynamics spanning a wide range of time scales.

physics.optics↗

arXiv2022

Chaotic mode-competition dynamics in a multimode semiconductor laser with optical feedback and injection

Photonic computing is attracting increasing interest to accelerate information processing in machine learning applications. The mode-competition dynamics of multimode semiconductor lasers is useful for solving the multi-armed bandit problem in reinforcement learning for computing applications. In this study, we numerically evaluate the chaotic mode-competition dynamics in a multimode semiconductor laser with optical feedback and injection. We observe the chaotic mode-competition dynamics among the longitudinal modes and control them by injecting an external optical signal into one of the longitudinal modes. We define the dominant mode as the mode with the maximum intensity; the dominant-mode ratio for the injected mode increases as the optical injection strength increases. We find that the characteristics of the dominant mode ratio in terms of the optical injection strength are different among the modes owing to the different optical feedback phases. We propose a control technique for the characteristics of the dominant mode ratio by precisely tuning the initial optical frequency detuning between the optical injection signal and injected mode. We also evaluate the relationship between the region for the large dominant mode ratio and injection locking range. The region for the large dominant mode ratio does not correspond to the injection-locking range. This discrepancy results from the complex mode-competition dynamics in multimode semiconductor lasers with both optical feedback and injection. This control technique of chaotic mode-competition dynamics in multimode lasers is promising for applications in reinforcement learning and reservoir computing as photonic artificial intelligence.

physics.optics↗

arXiv2022

Parallel photonic accelerator for decision making using optical spatiotemporal chaos

Photonic accelerators have attracted increasing attention in artificial intelligence applications. The multi-armed bandit problem is a fundamental problem of decision making using reinforcement learning. However, the scalability of photonic decision making has not yet been demonstrated in experiments, owing to technical difficulties in physical realization. We propose a parallel photonic decision-making system for solving large-scale multi-armed bandit problems using optical spatiotemporal chaos. We solve a 512-armed bandit problem online, which is much larger than previous experiments by two orders of magnitude. The scaling property for correct decision making is examined as a function of the number of slot machines, evaluated as an exponent of 0.86. This exponent is smaller than that in previous work, indicating the superiority of the proposed parallel principle. This experimental demonstration facilitates photonic decision making to solve large-scale multi-armed bandit problems for future photonic accelerators.

cs.ET↗