Scientific research increasingly relies on distributed computational resources, storage systems, networks, and instruments, ranging from HPC and cloud systems to edge devices. Event-driven architecture (EDA) benefits ...
详细信息
The InterPlanetary File System (IPFS) is on its way to becoming the backbone of the next generation of the web. However, it suffers from several performance bottlenecks, particularly on the content retrieval path, whi...
详细信息
Sensor-cloud systems (SCSs) aim to provide flexible configurable platforms for monitoring and control-ling the IoT-enabled applications. By integrating sensors, wireless networks and cloud for managing sensors, collec...
详细信息
Sensor-cloud systems (SCSs) aim to provide flexible configurable platforms for monitoring and control-ling the IoT-enabled applications. By integrating sensors, wireless networks and cloud for managing sensors, collecting data, and automating decision-making, the collected sensing data are typically used for machine learning purposes. With increasing emphasis in privacy protection, Federated Learning (FL) is widely adopted for enhancing privacy preservation. FL enables sharing of data for machine learning while preserving the privacy of the data owners. In SCSs, FL involves a large number of edge nodes in order to ensure a sufficient amount of data for model training. However, FL inevitably incurs prohibitive overheads if it simply gathers data from all the nodes, hence making it desirable to adopt some scheduling strategy so that data are collected only from a selected subset of nodes. This paper proposes a scheduling strategy based on deep reinforcement learning (DRL) for improving the performance and efficiency of FL in SCSs. The DRL environment, such as state space, action space, and reward function, is carefully designed. Proximal policy optimization is employed to train the DRL agent. Experimental results demonstrated that the proposed method outperforms other baselines on both independent and identically distributed (IID) and non-IID datasets.(c) 2023 Elsevier B.V. All rights reserved.
Lattice-based Post-Quantum Cryptography (PQC) can effectively resist the quantum threat to blockchain's underlying cryptographic algorithms. Blockchain node decryption is one of the most commonly used cryptographi...
详细信息
The increasing demand for radio spectrum, spurred by an upsurge in interconnected devices, necessitates innovative management solutions. Cognitive Radio networks (CRN) offer a promising approach to dynamically allocat...
详细信息
Object detection in remote sensing images is a significant and challenging task. The detection performance is difficult to further improve due to the wide variation in object scale and unpredictable orientations. Howe...
详细信息
With the rapid development of VAV technology and the Internet of Things (IoT), the application of VAVs in emergency scenarios has become a research hotspot. In emergency situations, how to efficiently plan UAV flight ...
详细信息
Generative AI, a novel general-purpose technology, has the capability to produce outputs closely resembling its training inputs. This paper presents the first use of such learning in memory workload synthesis. By devi...
详细信息
ISBN:
(纸本)9798400716447
Generative AI, a novel general-purpose technology, has the capability to produce outputs closely resembling its training inputs. This paper presents the first use of such learning in memory workload synthesis. By devising three learning techniques and comparing them with two existing techniques, the paper shows that by carefully choosing the training input and processing the generated output, AI-based synthesis can produce generated workloads that have similar or better accuracy than the best of existing methods.
The proceedings contain 853 papers. The topics discussed include: design and analysis of an energy-efficient duo-core SRAM-based compute-in-memory accelerator;effective resource model and cost scheme for maze routing ...
ISBN:
(纸本)9798350330991
The proceedings contain 853 papers. The topics discussed include: design and analysis of an energy-efficient duo-core SRAM-based compute-in-memory accelerator;effective resource model and cost scheme for maze routing in 3D global routing;SPRCPl: an efficient tool for SNN models deployment on multi-core neuromorphic chips via pilot running;a lumped circuit model for implantable body-coupled channel;a CMOS wideband linear low-noise amplifier using dual capacitor-cross-coupled configurations;audio-visual cross-modal generation with multimodal variational generative model;hybrid event-frame neural spike detector for neuromorphic implantable BMI;a multi-stride convolution acceleration algorithm for CNNs;and analysis for optimizer based on spiking-neural oscillator networks with a simple network topology.
Multiple input multiple output (MIMO) plays an important role in realizing the present fifth-generation (5G) as well as the next-generation wireless networks. The channel estimation process in the majority of the exis...
详细信息
暂无评论