Lyzis Labs is an incentive-driven and democratic protocol built upon a decentralized online marketplace based on blockchain technology. The major proposal of the protocol, Lyzis Marketplace, allows two or more people ...
详细信息
ISBN:
(纸本)9781665464598
Lyzis Labs is an incentive-driven and democratic protocol built upon a decentralized online marketplace based on blockchain technology. The major proposal of the protocol, Lyzis Marketplace, allows two or more people to be securely connected in a decentralized way without going through a Trusted Third Party (TTP) to perform physical asset exchanges while primarily providing transparent and fully protected data storage. The necessary components are built to ensure that within the platform, the honest strategy of an actor (seller or buyer) is safe, in a strong game-theoretic sense, if the arbiter is biased in favor of the honest parties. Such approach may give rise to a permissionless, secure and transparent business system where users are empowered by decision-making and can simultaneously follow personal and collective interests.
This paper presents a centralized model based on metaheuristics to solve the problem of optimal Electric Vehicle (EV) charge scheduling in multiple parking lots. A centralized optimization model using two-level partic...
详细信息
distributed Ledger Technology (DLT) is a decentralized database system where transactions are recorded and verified across multiple nodes. Its key features include immutability, time-stamping, and consensus-based vali...
详细信息
This paper delves into the realm of quantum computing and its potential to revolutionize data encryption methodologies. Leveraging IBM's Qiskit tool, we investigate encryption approaches aimed at bolstering data s...
详细信息
Unmanned aerial vehicles (UAVs) are used as supportive edge computing for sparsely located user equipment on a large scale. In this work, we propose and address a collaborative edge computing system involving multiple...
详细信息
The proceedings contain 8 papers. The topics discussed include: detecting and defending vulnerabilities in heterogeneous and monolithic systems: current strategies and future directions;enabling energy-efficient AI co...
ISBN:
(纸本)9798350356373
The proceedings contain 8 papers. The topics discussed include: detecting and defending vulnerabilities in heterogeneous and monolithic systems: current strategies and future directions;enabling energy-efficient AI computing: leveraging application-specific approximations;efficient neural networks: from SW optimization to specialized HW accelerators;what do transformers have to learn from biological spiking neural networks?;primer on data in quantum machine learning;work-in-progress: ACPO: an AI-enabled compiler framework;and work-in-progress: temporal RegionDrop - frame difference sparsity for efficient video inference.
This paper focuses on the design and analysis of silicon photonic waveguides and optical micro-ring resonators, with a vision to advance next-generation computing, communication, and sensing technologies. Silicon phot...
详细信息
Noisy Intermediate-Scale Quantum (NISQ) computers currently available have a few thousand qubits, and could potentially solve combinatorial optimization problems efficiently. However, the sizes of the problems that co...
详细信息
ISBN:
(纸本)9798331541378
Noisy Intermediate-Scale Quantum (NISQ) computers currently available have a few thousand qubits, and could potentially solve combinatorial optimization problems efficiently. However, the sizes of the problems that could be solved are limited by the number of qubits, their connectivity, high noise, and short coherence times. In this work, we propose hybrid quantum-classical algorithms based on the divide and conquer paradigm for solving larger Maximum Independent Set (MIS) and Maximum Weighted Independent Set (MWIS) problems on graphs than would otherwise be possible on NISQ devices. The machines include D-Wave Quantum Annealers and QuEra Quantum Computers with Neutral Atoms. Our algorithms are designed for separable graphs, which are classes of graphs with good vertex separators;these include planar graphs, finite element meshes with good aspect ratios, nearest neighbor graphs, and certain classes of geometrically defined graphs. The algorithms recursively employ vertex separators to create a collection of small subgraphs, from which NISQ computers obtain disjoint maximal independent sets, which are then augmented by independent vertices from the separators. We demonstrate competitive results compared to the classical Luby's algorithm, and KaMIS, a state-of-the-art classical MIS solver, on graphs with several thousand vertices. These divide and conquer-based algorithms are also well-suited for distributed quantum computer architectures.
This paper presents novel hierarchical management architecture of heterogeneous edge data center. The management architecture was implemented in BRAINE (Big data pRocessing and Artificial Intelligence at the Network E...
详细信息
Partitioning and deploying Deep Neural Networks (DNNs) across edge nodes may be used to meet performance objectives of applications. However, the failure of a single node may result in cascading failures that will adv...
详细信息
ISBN:
(数字)9781665481403
ISBN:
(纸本)9781665481403
Partitioning and deploying Deep Neural Networks (DNNs) across edge nodes may be used to meet performance objectives of applications. However, the failure of a single node may result in cascading failures that will adversely impact the delivery of the service and will result in failure to meet specific objectives. The impact of these failures needs to be minimised at runtime. Three techniques are explored in this paper, namely repartitioning, early-exit and skip-connection. When an edge node fails, the repartitioning technique will repartition and redeploy the DNN thus avoiding the failed nodes. The earlyexit technique makes provision for a request to exit (early) before the failed node. The skip connection technique dynamically routes the request by skipping the failed nodes. This paper will leverage trade-offs in accuracy, end-to-end latency and downtime for selecting the best technique given user-defined objectives (accuracy, latency and downtime thresholds) when an edge node fails. To this end, CONTINUER is developed. Two key activities of the framework are estimating the accuracy and latency when using the techniques for distributed DNNs and selecting the best technique. It is demonstrated on a lab-based experimental testbed that CONTINUER estimates accuracy and latency when using the techniques with no more than an average error of 0.28% and 13.06%, respectively, and selects the suitable technique with a low overhead of no more than 16.82 milliseconds and an accuracy of up to 99.86%.
暂无评论