相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Optics
Alexander D. White et al.
Summary: This article demonstrates an integrated approach for passively isolating a continuous-wave laser using the non-reciprocal Kerr nonlinearity in ring resonators. By using silicon nitride as the model platform, the authors achieve single ring isolation of 17-23 dB with 1.8-5.5 dB insertion loss, and cascaded ring isolation of 35 dB with 5 dB insertion loss. They also demonstrate hybrid integration and isolation with a semiconductor laser chip using these devices.
Article
Multidisciplinary Sciences
Seungchul Jung et al.
Summary: This paper presents a 64 x 64 crossbar array based on MRAM cells that overcomes the low-resistance issue and successfully implements analogue multiply-accumulate operations in artificial neural networks. The researchers achieved high classification accuracy and face detection using this array for tasks involving 10,000 digits.
Review
Optics
Hailong Zhou et al.
Summary: Photonic matrix multiplication is a crucial part of information processing in science and technology, offering faster and more efficient computations in the optical domain. Recent research has shown its potential for applications in telecommunications and artificial intelligence that are beyond the capabilities of conventional processors.
LIGHT-SCIENCE & APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Changming Wu et al.
Summary: Researchers demonstrate a photonic generative network that can generate handwritten numbers and show resilience to hardware nonidealities. These results suggest the potential of more complex photonic generative networks based on photonic hardware.
Article
Multidisciplinary Sciences
Bowei Dong et al.
Summary: This study demonstrates biometrics-protected optical communication through synergizing triboelectric and nanophotonic technology. The method allows for multiplexing of digital and biometric information with zero power consumption, and achieves successful user identification and high-speed digital information recovery, enabling secure communication and smart home control.
Article
Multidisciplinary Sciences
Haowen Shu et al.
Summary: This research combines microcomb and SiPh technologies to achieve a low-cost and efficient integrated photonics system, providing new solutions in the fields of optical data transmission and microwave photonics.
Article
Multidisciplinary Sciences
Yannis Assael et al.
Summary: The study introduces Ithaca, a deep neural network for restoring, attributing, and dating ancient Greek inscriptions. The use of Ithaca significantly improves the accuracy of text restoration and attribution compared to historians working alone, contributing to the study of ancient history.
Article
Multidisciplinary Sciences
Farshid Ashtiani et al.
Summary: This study presents an integrated photonic deep neural network that performs image classification with sub-nanosecond classification time. Optical waves are directly processed to achieve linear computation in each neuron and opto-electronically realized non-linear activation functions, resulting in comparable computation speed and scalability with digital platforms.
Article
Nanoscience & Nanotechnology
Syed Ghazi Sarwat et al.
Summary: Phase-change memtransistive synapses with tunable plasticities are introduced, combining the non-volatility of phase configurations and volatility of field-effect modulation. These synapses can model dynamic environments and accelerate the solving of combinatorial optimization problems in Hopfield neural networks.
NATURE NANOTECHNOLOGY
(2022)
Review
Multidisciplinary Sciences
Mario Lanza et al.
Summary: Memristive devices, which can change their resistance and memory state, have potential applications in various fields. However, there are still challenges to be addressed, including performance and reliability issues.
Article
Multidisciplinary Sciences
Min-Kyu Kim et al.
Summary: In this study, integrated ferroelectric thin-film transistor (FeTFT) synaptic arrays were demonstrated to provide efficient parallel programming and data processing for CNNs through selective and accurate control of polarization in the ferroelectric layer.
Article
Multidisciplinary Sciences
Weier Wan et al.
Summary: This study presents NeuRRAM, a RRAM-based CIM chip that offers versatility, high energy efficiency, and accuracy. By co-optimizing algorithms, architecture, circuits, and devices, the chip can be reconfigured for different model architectures and provides twice the energy efficiency of previous state-of-the-art RRAM-CIM chips across various computational bit-precisions. It achieves inference accuracy comparable to software models quantized to four-bit weights across various AI tasks.
Article
Multidisciplinary Sciences
Yang Liu et al.
Summary: Erbium-doped fiber amplifiers have revolutionized optical communications and laser technology, but their practical use is limited by insufficient output power. This study demonstrates a photonic integrated circuit-based erbium amplifier that achieves high output power and gain, surpassing state-of-the-art semiconductor amplifiers. By applying ion implantation to low-loss silicon nitride photonic integrated circuits, various fiber-based devices can be miniaturized.
Article
Multidisciplinary Sciences
June Sang Lee et al.
Summary: This study introduces hybridized-active-dielectric nanowires for polarization-selective tunability and demonstrates the ability to modulate conductance and perform matrix-vector multiplication using polarization as a parameter. This concept can be generalized to other materials and has potential in various applications.
Article
Multidisciplinary Sciences
Tao Yan et al.
Summary: Photonic neural networks use photons instead of electrons to perform brain-like computations, leading to significantly improved computing performance. However, current architectures are limited to handling data with regular structures and cannot generalize to graph-structured data beyond Euclidean space. In this study, a diffractive graph neural network (DGNN) is proposed to address this limitation by utilizing diffractive photonic computing units (DPUs) and on-chip optical devices. DGNN achieves complex feature representation by capturing dependencies among node neighborhoods during light-speed optical message passing over graph structures. It demonstrates superior performance in node and graph-level classification tasks with benchmark databases, providing a new direction for high-efficiency processing of large-scale graph data structures using deep learning.
Article
Multidisciplinary Sciences
Minh A. Trail et al.
Summary: Integrated photonics has greatly impacted various technologies in modern society, offering scalability, weight, cost and power efficiency. We present a new generation of integrated photonics by combining III-V materials with silicon nitride waveguides, enabling the fabrication of fully integrated photonic integrated circuits (PICs) operating at submicrometre wavelengths. This integration strategy unlocks a broad range of new photonics applications, with exceptional performance at short wavelengths.
Article
Multidisciplinary Sciences
Alhussein Fawzi et al.
Summary: Improving the efficiency of fundamental computations, such as matrix multiplication, is crucial for various fields, and machine learning can automate the discovery of efficient algorithms that surpass human-designed ones. AlphaTensor, based on deep reinforcement learning and AlphaZero, successfully discovered algorithms that outperform state-of-the-art ones. Its flexibility in accelerating algorithm discovery and optimization is showcased through different use-cases.
Article
Multidisciplinary Sciences
Ziyuan Rao et al.
Summary: This study proposes an active learning strategy to accelerate the design of high-entropy Invar alloys. By integrating machine learning with density-functional theory, thermodynamic calculations, and experiments, the researchers successfully identified high-entropy Invar alloys with extremely low thermal expansion coefficients. This approach shows promise for the fast and automated discovery of high-entropy alloys with optimal thermal, magnetic, and electrical properties.
Article
Multidisciplinary Sciences
J. Dauparas et al.
Summary: This article introduces a deep learning-based protein sequence design method, ProteinMPNN, which has shown outstanding performance in both theoretical and experimental tests, suggesting its wide applicability.
Article
Multidisciplinary Sciences
Alexander Sludds et al.
Article
Multidisciplinary Sciences
J. Feldmann et al.
Summary: With the advancement of technology, the demand for fast processing of large amounts of data is increasing, making highly parallelized, fast, and scalable hardware crucial. The integration of photonics can serve as the optical analogue of an application-specific integrated circuit, enabling photonic in-memory computing and efficient computational hardware.
Article
Multidisciplinary Sciences
Xingyuan Xu et al.
Summary: Inspired by biological visual cortex systems, convolutional neural networks extract hierarchical features of raw data, reducing parameter complexity and improving prediction accuracy. Optical neural networks promise faster computing using broad optical bandwidths, with optical vector convolutional accelerators demonstrating efficient image processing capabilities.
Review
Optics
Bhavin J. Shastri et al.
Summary: Research in photonic computing is thriving due to the proliferation of optoelectronic components in photonic integration platforms. Photonic integrated circuits have enabled ultrafast artificial neural networks, with algorithms running on such hardware having the potential to address the growing demand for machine learning and artificial intelligence. Neuromorphic photonics offers sub-nanosecond latencies, providing a complementary opportunity to extend the domain of artificial intelligence.
Article
Nanoscience & Nanotechnology
Cong Wang et al.
Summary: The growth of intelligent devices in the Internet of Things has created a need for real-time processing of large volumes of analogue data. Utilizing continuous data representation and frequency multiplexing in nanoscale crossbar arrays enables the implementation of a scalable massively parallel computing scheme for direct processing of analogue information.
NATURE NANOTECHNOLOGY
(2021)
Article
Psychology, Biological
Kiyohito Iigaya et al.
Summary: The study found that aesthetic preferences for visual art can be predicted by a model that combines low-level and high-level image features, and that a convolutional neural network trained only on object recognition naturally encodes these features.
NATURE HUMAN BEHAVIOUR
(2021)
Proceedings Paper
Computer Science, Hardware & Architecture
Albert Reuther et al.
Summary: This paper provides an update on the survey of AI accelerators and processors from the past two years, collecting and summarizing current commercial accelerators with peak performance and power consumption numbers. It also analyzes trends and computational efficiency in terms of peak performance.
2021 IEEE HIGH PERFORMANCE EXTREME COMPUTING CONFERENCE (HPEC)
(2021)
Article
Engineering, Electrical & Electronic
Mitchell A. Nahmias et al.
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS
(2020)
Article
Multidisciplinary Sciences
Peng Yao et al.
Review
Nanoscience & Nanotechnology
Abu Sebastian et al.
NATURE NANOTECHNOLOGY
(2020)
Review
Multidisciplinary Sciences
Gordon Wetzstein et al.
Summary: Artificial intelligence tasks require accelerators for fast and low-power execution. While general-purpose optical computing systems have not matured into a practical technology, recent research shows promise for optical computing in artificial intelligence applications, particularly for visual computing. The potential and challenges of all-optical and hybrid optical networks are discussed in this Perspective.
Article
Multidisciplinary Sciences
Carlos Rios et al.
Article
Multidisciplinary Sciences
Cong Han et al.
Review
Optics
Luqi Yuan et al.
Article
Optics
Yichen Shen et al.
Article
Computer Science, Artificial Intelligence
Lina Zhou et al.
Article
Optics
Weilin Liu et al.
Review
Computer Science, Information Systems
Weisong Shi et al.
IEEE INTERNET OF THINGS JOURNAL
(2016)
Proceedings Paper
Computer Science, Hardware & Architecture
Ikuo Magaki et al.
2016 ACM/IEEE 43RD ANNUAL INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE (ISCA)
(2016)
Proceedings Paper
Computer Science, Theory & Methods
Xiaqing Li et al.
PROCEEDINGS 45TH INTERNATIONAL CONFERENCE ON PARALLEL PROCESSING - ICPP 2016
(2016)
Review
Multidisciplinary Sciences
Yann LeCun et al.
Article
Optics
Carlos Rios et al.
Article
Engineering, Electrical & Electronic
Alexander N. Tait et al.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2014)
Review
Multidisciplinary Sciences
Igor L. Markov
Article
Instruments & Instrumentation
M. T. Baig et al.
REVIEW OF SCIENTIFIC INSTRUMENTS
(2013)
Article
Engineering, Electrical & Electronic
Hua Ji et al.
IEEE PHOTONICS TECHNOLOGY LETTERS
(2010)