Computational Network Research Articles

Dynamics and resonance fluorescence from a superconducting artificial atom doubly driven by quantized and classical fields

We report an experimental demonstration of resonance fluorescence in a two-level superconducting artificial atom under two driving fields coupled to a detuned cavity. One of the fields is classical and the other is varied from quantum (vacuum fluctuations) to classical by controlling the photon number inside the cavity. The device consists of a transmon qubit strongly coupled to a one-dimensional transmission line and a coplanar waveguide resonator. We observe a sideband anticrossing and asymmetry in the emission spectra of the system through a one-dimensional transmission line, which is fundamentally different from the weak-coupling case. By changing the photon number inside the cavity, the emission spectrum of our doubly driven system approaches the case when the atom is driven by two classical bichromatic fields. We also measure the dynamical evolution of the system through the transmission line and study the properties of the first-order correlation function, Rabi oscillations, and energy relaxation in the system. The study of resonance fluorescence from an atom driven by two fields promotes understanding decoherence in superconducting quantum circuits and may find applications in superconducting quantum computing and quantum networks. Published by the American Physical Society 2024

An In-Memory-Computing Structure with Quantum-Dot Transistor Toward Neural Network Applications: From Analog Circuits to Memory Arrays

The rapid advancements in artificial intelligence (AI) have demonstrated great success in various applications, such as cloud computing, deep learning, and neural networks, among others. However, the majority of these applications rely on fast computation and large storage, which poses significant challenges to the hardware platform. Thus, there is a growing interest in exploring new computation architectures to address these challenges. Compute-in-memory (CIM) has emerged as a promising solution to overcome the challenges posed by traditional computer architecture in terms of data transfer frequency and energy consumption. Non-volatile memory, such as Quantum-dot transistors, has been widely used in CIM to provide high-speed processing, low power consumption, and large storage capacity. Matrix-vector multiplication (MVM) or dot product operation is a primary computational kernel in neural networks. CIM offers an effective way to optimize the performance of the dot product operation by performing it through an intertwining of processing and memory elements. In this paper, we present a novel design and analysis of a Quantum-dot transistor (QDT) based CIM that offers efficient MVM or dot product operation by performing computations inside the memory array itself. Our proposed approach offers energy-efficient and high-speed data processing capabilities that are critical for implementing AI applications on resource-limited platforms such as portable devices.

The Collaboverse: A Collaborative Data-Sharing and Speech Analysis Platform.

Collaboration in the field of speech-language pathology occurs across a variety of digital devices and can entail the usage of multiple software tools, systems, file formats, and even programming languages. Unfortunately, gaps between the laboratory, clinic, and classroom can emerge in part because of siloing of data and workflows, as well as the digital divide between users. The purpose of this tutorial is to present the Collaboverse, a web-based collaborative system that unifies these domains, and describe the application of this tool to common tasks in speech-language pathology. In addition, we demonstrate its utility in machine learning (ML) applications. This tutorial outlines key concepts in the digital divide, data management, distributed computing, and ML. It introduces the Collaboverse workspace for researchers, clinicians, and educators in speech-language pathology who wish to improve their collaborative network and leverage advanced computation abilities. It also details an ML approach to prosodic analysis. The Collaboverse shows promise in narrowing the digital divide and is capable of generating clinically relevant data, specifically in the area of prosody, whose computational complexity has limited widespread analysis in research and clinic alike. In addition, it includes an augmentative and alternative communication app allowing visual, nontextual communication.

Physical reservoir computing and deep neural networks using artificial and natural noncollinear spin textures

Physical reservoir computing and deep neural networks using artificial and natural noncollinear spin textures

Coherent spin qubit shuttling through germanium quantum dots

Quantum links can interconnect qubit registers and are therefore essential in networked quantum computing. Semiconductor quantum dot qubits have seen significant progress in the high-fidelity operation of small qubit registers but establishing a compelling quantum link remains a challenge. Here, we show that a spin qubit can be shuttled through multiple quantum dots while preserving its quantum information. Remarkably, we achieve these results using hole spin qubits in germanium, despite the presence of strong spin-orbit interaction. In a minimal quantum dot chain, we accomplish the shuttling of spin basis states over effective lengths beyond 300 microns and demonstrate the coherent shuttling of superposition states over effective lengths corresponding to 9 microns, which we can extend to 49 microns by incorporating dynamical decoupling. These findings indicate qubit shuttling as an effective approach to route qubits within registers and to establish quantum links between registers.

Estimation of the Multifractal Spectrum Characteristics of Fractal Dimension of Network Traffic and Computer Attacks in IoT

Estimation of the Multifractal Spectrum Characteristics of Fractal Dimension of Network Traffic and Computer Attacks in IoT

The topology of local computing in networks

The topology of local computing in networks

Multi-layered computational gene networks by engineered tristate logics

So far, biocomputation strictly follows traditional design principles of digital electronics, which could reach their limits when assembling gene circuits of higher complexity. Here, by creating genetic variants of tristate buffers instead of using conventional logic gates as basic signal processing units, we introduce a tristate-based logic synthesis (TriLoS) framework for resource-efficient design of multi-layered gene networks capable of performing complex Boolean calculus within single-cell populations. This sets the stage for simple, modular, and low-interference mapping of various arithmetic logics of interest and an effectively enlarged engineering space within single cells. We not only construct computational gene networks running full adder and full subtractor operations at a cellular level but also describe a treatment paradigm building on programmable cell-based therapeutics, allowing for adjustable and disease-specific drug secretion logics invivo. This work could foster the evolution of modern biocomputers to progress toward unexplored applications in precision medicine.

Adapting Progressive Hedging for Solving Two-Stage Stochastic Programs Under a Peer-to-Peer Computing Network

Adapting Progressive Hedging for Solving Two-Stage Stochastic Programs Under a Peer-to-Peer Computing Network

Evaluating climate change impacts on snow cover and karst spring discharge in a data-scarce region: a case study of Iran

AbstractThe incremental impacts of climate change on elements within the water cycle are a growing concern. Intricate karst aquifers have received limited attention concerning climate change, especially those with sparse data. Additionally, snow cover has been overlooked in simulating karst spring discharge rates. This study aims to assess climate change effects in a data-scarce karst anticline, specifically Khorramabad, Iran, focusing on temperature, precipitation, snow cover, and Kio spring flows. Utilizing two shared socioeconomic pathways (SSPs), namely SSP2-4.5 and SSP5-8.5, extracted from the CMIP6 dataset for the base period (1991–2018) and future periods (2021–2040 and 2041–2060), the research employs Landsat data and artificial neural networks (ANNs) for snow cover and spring discharge computation, respectively. ANNs are trained using the training and verification periods of 1991–2010 and 2011–2018, respectively. Results indicate projected increases in temperature, between + 1.21 °C (2021–2040 under SSP245) and + 2.93 °C (2041–2060 under SSP585), and precipitation, from + 2.91 mm/month (2041–2060 under SSP585) to + 4.86 mm/month (2021–2040 under SSP585). The ANN models satisfactorily simulate spring discharge and snow cover, predicting a decrease in snow cover between − 4 km2/month (2021–2040 under SSP245) and − 11.4 km2/month (2041–2060 under SSP585). Spring discharges are anticipated to increase from + 28.5 l/s (2021–2040 under SSP245) to + 57 l/s (2041–2060 under SSP585) and from + 12.1 l/s (2021–2040 under SSP585) to + 36.1 l/s (2041–2060 under SSP245), with and without snow cover as an input, respectively. These findings emphasize the importance of considering these changes for the sustainability of karst groundwater in the future.

Joint Communication-Caching-Computing Resource Allocation for Bidirectional Data Computation in IRS-Assisted Hybrid UAV-Terrestrial Network

Joint Communication-Caching-Computing Resource Allocation for Bidirectional Data Computation in IRS-Assisted Hybrid UAV-Terrestrial Network

Flexible multitask computation in recurrent networks utilizes shared dynamical motifs

Flexible computation is a hallmark of intelligent behavior. However, little is known about how neural networks contextually reconfigure for different computations. In the present work, we identified an algorithmic neural substrate for modular computation through the study of multitasking artificial recurrent neural networks. Dynamical systems analyses revealed learned computational strategies mirroring the modular subtask structure of the training task set. Dynamical motifs, which are recurring patterns of neural activity that implement specific computations through dynamics, such as attractors, decision boundaries and rotations, were reused across tasks. For example, tasks requiring memory of a continuous circular variable repurposed the same ring attractor. We showed that dynamical motifs were implemented by clusters of units when the unit activation function was restricted to be positive. Cluster lesions caused modular performance deficits. Motifs were reconfigured for fast transfer learning after an initial phase of learning. This work establishes dynamical motifs as a fundamental unit of compositional computation, intermediate between neuron and network. As whole-brain studies simultaneously record activity from multiple specialized systems, the dynamical motif framework will guide questions about specialization and generalization.

Semiconductor lasers for photonic neuromorphic computing and photonic spiking neural networks: A perspective

Photonic neuromorphic computing has emerged as a promising avenue toward building a high-speed, low-latency, and energy-efficient non-von-Neumann computing system. Photonic spiking neural network (PSNN) exploits brain-like spatiotemporal processing to realize high-performance neuromorphic computing. Linear weighting and nonlinear spiking activation are two fundamental functions of a SNN. However, the nonlinear computation of PSNN remains a significant challenge. Therefore, this perspective focuses on the nonlinear computation of photonic spiking neurons, including numerical simulation, device fabrication, and experimental demonstration. Different photonic spiking neurons are considered, such as vertical-cavity surface-emitting lasers, distributed feedback (DFB) lasers, Fabry–Pérot (FP) lasers, or semiconductor lasers embedded with saturable absorbers (SAs) (e.g., FP-SA and DFB-SA). PSNN architectures, including fully connected and convolutional structures, are developed, and supervised and unsupervised learning algorithms that take into account optical constraints are introduced to accomplish specific applications. This work covers devices, architectures, learning algorithms, and applications for photonic and optoelectronic neuromorphic computing and provides our perspective on the challenges and prospects of photonic neuromorphic computing based on semiconductor lasers.

Special Issue: Selected papers from the AIxIA 2023 Workshops

The 2023 edition of the AIxIA Conference, held in Rome, brought together a large number of researchers and practitioners to discuss the most recent and important advancements in Artificial Intelligence (AI). The conference featured 19 workshops, organized by 77 experts, attracting 248 submissions and resulting in 16 proceedings. This special issue presents extended versions of selected papers initially showcased at these workshops. Each paper underwent rigorous review and represents a diverse array of topics, reflecting the multifaceted nature of the Italian AI community. The topics covered include ethical foundations to symbiotic AI, symbolic knowledge extraction from black-box models, creative influence prediction using graph theory, AI approaches to multidimensional poverty prediction, an assessment of AI-based supports for informal caregivers, deep learning-based EEG denoising, AI-assisted board-game-based learning, large language models for assessment and feedback in higher education, geometric reasoning in the Traveling Salesperson Problem, defeasible reasoning in weighted knowledge bases, and conditional computation in neural networks. These contributions demonstrate the innovative and interdisciplinary research within the AI community, offering valuable insights and advancing the field.

Conditional computation in neural networks: Principles and research trends

This article summarizes principles and ideas from the emerging area of applying conditional computation methods to the design of neural networks. In particular, we focus on neural networks that can dynamically activate or de-activate parts of their computational graph conditionally on their input. Examples include the dynamic selection of, e.g., input tokens, layers (or sets of layers), and sub-modules inside each layer (e.g., channels in a convolutional filter). We first provide a general formalism to describe these techniques in an uniform way. Then, we introduce three notable implementations of these principles: mixture-of-experts (MoEs) networks, token selection mechanisms, and early-exit neural networks. The paper aims to provide a tutorial-like introduction to this growing field. To this end, we analyze the benefits of these modular designs in terms of efficiency, explainability, and transfer learning, with a focus on emerging applicative areas ranging from automated scientific discovery to semantic communication.

Development and Management of E-Commerce Information Systems Using Edge Computing and Neural Networks

E-commerce refers to implementing e-commerce on mobile devices, cell phones, and mobile Internet connections. The fast advancement of mobile electronic gadgets and mobile Internet connections has facilitated the emergence of this new e-commerce market. The e-commerce systems contain several concealed client behavior data and forthcoming growth patterns. Data mining technologies extract valuable data and facilitate the growth of e-commerce. This article focuses on studying mobile e-commerce systems within the framework of edge computing, specifically in the setting of industrial clusters. This study examines the importance and benefits of data mining techniques in implementing e-commerce management platforms using neural networks. It also evaluates the corresponding methods of data mining and predicts purchasing trends. The study has used the benefits of grouping and back propagation neural network approaches in data analytics to categorize goods details, buying tastes, and related data. The advantage of neural networks in processing nonlinear patterns is employed to forecast forthcoming buying power. The findings demonstrate that data analytical methods and neural networks exhibit higher precision in predicting the buying pattern. The correlating factor between actual and projected usage information was 0.98, with the highest relative average inaccuracy of 2.4%. Data mining technologies effectively extract previously unrecognized relevant data and anticipated buying patterns in e-commerce platforms. Neural networks have a solid ability to predict future consumption capacity and consumption behaviors accurately.

Design of an information-computer system for the automated modeling of systems for automatic orientation of production objects in the machine and instrument industries

An information-computer system has been developed for the automated modeling of systems for automatic orientation of production objects, which is one of the most important and complex creative flexible production systems of machine and instrument engineering. The proposed information-computer system for automated modeling of systems of automatic orientation of production objects is an effective tool for solving an important task of a scientific and applied nature. Its use makes it possible to increase the speed and efficiency of information processing and to make correct and well-founded decisions when determining the composition and method of organization of systems of automatic orientation of production objects. The structure of this information-computer system is a specific set of software and hardware and information and telecommunication tools and interactive functional modules. This structure reproduces a certain paradigm that conditions the integrity and integration of the information-computer system for automated modeling of systems for automatic orientation of production objects in flexible production systems. In addition, uniformity, extensibility, the possibility of modernization and changeability of software components, protection of information from unauthorized access and preservation of commercial secrets are ensured according to international criteria for evaluating the protection of the computer system. Neuro-fuzzy network information processing and computer vision algorithms have been implemented for automatic identification of production objects and orientation devices, which are components of automatic orientation systems of production objects. The developed information-computer system processes information in real time with high accuracy and speed

Comparative analysis of single-board computers for the development of a microarchitectural computing system for fire detection

Objectives. The purpose of the work is to select the basic computing microplatform of the onboard microarchitectural computing complex for the detection of anomalous situations in the territory of the Republic of Belarus from space on the basis of artificial intelligence methods.Methods. The method of comparative analysis is used to select a computing platform. A series of performance tests and comparative analysis (benchmarking) are performed on the selected equipment. The methods of comparative and benchmarking analysis are performed in accordance with the terms of reference to the current project.Results. A comparative analysis and performance testing of Raspberry Pi 4 Model B and Cool Pi 4 Model B single-board computers, as well as AI-accelerator Google Coral USB Accelerator with Google Edge TPU have been performed. The comparative analysis showed that Raspberry Pi 4 Model B and Cool Pi 4 Model B fully meet the terms of reference to the current project. At the same time Cool Pi 4 Model B handles neural network calculations well, but four times slower than similar calculations on Google Coral USB Accelerator. Neural network computations on the Raspberry Pi 4 Model B are 22 times slower than similar computations on the Google Coral USB Accelerator. Cool Pi 4 Model B outperforms Raspberry Pi 4 Model B by the factor of two to three for data copying and compression and almost six times faster for neural network computations.Conclusion. Despite the fact that Raspberry Pi 4 Model B meets the terms of reference to the project as a computational basis, when developing an on-board microarchitectural computing system for detecting anomalous situations, it is worth using more powerful alternatives with built-in AI-accelerators (e.g., Radxa Rock 5 Model A) or with an additional external AI-accelerator (e.g., a combination of Cool Pi 4 Model B and Google Coral USB Accelerator). Using a Raspberry Pi 4 Model B with an additional AI-accelerator is also acceptable and will speed up computations by several dozen times. AI-accelerators provide the fastest neural network computations, but there are features related to the novelty of the technology that will be explored in further development.

Performance on massive MIMO enabled mobile edge computing networks: Parallel computing modeling

With the development of communication and computation technologies, mobile edge computing (MEC) has become an emerging technology being deployed in 5G/B5G mobile networks to improve the mobile user experience. In this paper, we study the asymptotic performance of parallel computing in MEC networks with massive multi-input and multi-output (MIMO) in the millimeter wave band. First, a massive MIMO enabled MEC network with a parallel-limited computing model is constructed, and the task completed probability is defined as a new network performance metric. In the proposed model, we derive the task offloading and computing probabilities via the probability generation functional of the Poisson point process, from which, the task completed probability is obtained. Then we propose a computing resource allocation optimization algorithm to maximize the task completed probability within a given delay constraint. The numerical simulation results prove the accuracy of the obtained theoretical results and verify the performance of the proposed algorithm.

Chemical reservoir computation in a self-organizing reaction network

Chemical reaction networks, such as those found in metabolism and signalling pathways, enable cells to process information from their environment1,2. Current approaches to molecular information processing and computation typically pursue digital computation models and require extensive molecular-level engineering3. Despite considerable advances, these approaches have not reached the level of information processing capabilities seen in living systems. Here we report on the discovery and implementation of a chemical reservoir computer based on the formose reaction4. We demonstrate how this complex, self-organizing chemical reaction network can perform several nonlinear classification tasks in parallel, predict the dynamics of other complex systems and achieve time-series forecasting. This in chemico information processing system provides proof of principle for the emergent computational capabilities of complex chemical reaction networks, paving the way for a new class of biomimetic information processing systems.