Gustavo de Veciana
YOU?
Author Swipe
View article: Optimal Scheduling Algorithms for LLM Inference: Theory and Practice
Optimal Scheduling Algorithms for LLM Inference: Theory and Practice Open
With the growing use of Large Language Model (LLM)-based tools like ChatGPT, Perplexity, and Gemini across industries, there is a rising need for efficient LLM inference systems. These systems handle requests with a unique two-phase comput…
View article: Batching-Aware Joint Model Onloading and Offloading for Hierarchical Multi-Task Inference
Batching-Aware Joint Model Onloading and Offloading for Hierarchical Multi-Task Inference Open
The growing demand for intelligent services on resource-constrained edge devices has spurred the development of collaborative inference systems that distribute workloads across end devices, edge servers, and the cloud. While most existing …
View article: Importance Sampling via Score-based Generative Models
Importance Sampling via Score-based Generative Models Open
Importance sampling, which involves sampling from a probability density function (PDF) proportional to the product of an importance weight function and a base PDF, is a powerful technique with applications in variance reduction, biased or …
View article: Ten Ways in which Virtual Reality Differs from Video Streaming
Ten Ways in which Virtual Reality Differs from Video Streaming Open
Virtual Reality (VR) applications have a number of unique characteristics that set them apart from traditional video streaming. These characteristics have major implications on the design of VR rendering, adaptation, prefetching, caching, …
View article: Scheduling "Last Minute" Updates for Timely Decision-Making
Scheduling "Last Minute" Updates for Timely Decision-Making Open
We consider a setting where requests for updates regarding time-varying processes are required prior to making a sequence of decisions. Each request has a finite length time window during which the update should be received. The end of the…
View article: Managing Edge Offloading for Stochastic Workloads with Deadlines
Managing Edge Offloading for Stochastic Workloads with Deadlines Open
Increasing demand for computationally intensive jobs on mobile devices is driving interest in computation offloading to the edge/cloud servers. This paper presents a comprehensive framework for managing offloading of stochastic and heterog…
View article: Authors
Authors Open
To Re-transmit or Not to Re-transmit for Freshness WMOSC.4 635 Age-Based Cache Updating Under Timestomping Bang, Ban S4.4 111 An improved genetic algorithm for bi-level multi-objective Q-coverage in directional sensor networks Baras, John …
View article: MOHAWK: Mobility and Heterogeneity-Aware Dynamic Community Selection for Hierarchical Federated Learning
MOHAWK: Mobility and Heterogeneity-Aware Dynamic Community Selection for Hierarchical Federated Learning Open
The recent developments in Federated Learning (FL) focus on optimizing the learning process for data, hardware, and model heterogeneity. However, most approaches assume all devices are stationary, charging, and always connected to the Wi-F…
View article: Constrained Network Slicing Games: Achieving Service Guarantees and Network Efficiency
Constrained Network Slicing Games: Achieving Service Guarantees and Network Efficiency Open
Network slicing is a key capability for next generation mobile networks. It enables infrastructure providers to cost effectively customize logical networks over a shared infrastructure. A critical component of network slicing is resource a…
View article: Network Adaptive Federated Learning: Congestion and Lossy Compression
Network Adaptive Federated Learning: Congestion and Lossy Compression Open
In order to achieve the dual goals of privacy and learning across distributed data, Federated Learning (FL) systems rely on frequent exchanges of large files (model updates) between a set of clients and the server. As such FL systems are e…
View article: Online learning for multi-agent based resource allocation in weakly coupled wireless systems
Online learning for multi-agent based resource allocation in weakly coupled wireless systems Open
We propose and evaluate a learning-based framework to address multi-agent resource allocation in coupled wireless systems. In particular we consider, multiple agents (e.g., base stations, access points, etc.) that choose amongst a set of r…
View article: Performance and efficiency tradeoffs in blockchain overlay networks
Performance and efficiency tradeoffs in blockchain overlay networks Open
Underlying blockchain's scalability and performance is a Peer-to-Peer (P2P) overlay network and protocols for relaying blocks and transactions among participating nodes. In this work, we model and perform a systematic analysis of blockchai…
View article: Federated Learning Under Intermittent Client Availability and Time-Varying Communication Constraints
Federated Learning Under Intermittent Client Availability and Time-Varying Communication Constraints Open
Federated learning systems facilitate training of global models in settings where potentially heterogeneous data is distributed across a large number of clients. Such systems operate in settings with intermittent client availability and/or…
View article: Joint Scheduling of URLLC and eMBB Traffic in 5G Wireless Networks
Joint Scheduling of URLLC and eMBB Traffic in 5G Wireless Networks Open
Emerging 5G systems will need to efficiently support both enhanced mobile broadband traffic (eMBB) and ultra-low-latency communications (URLLC) traffic. In these systems, time is divided into slots which are further sub-divided into minisl…
View article: Book-Ahead & Supply Management for Ridesourcing Platforms
Book-Ahead & Supply Management for Ridesourcing Platforms Open
Ridesourcing platforms recently introduced the ``schedule a ride'' service where passengers may reserve (book-ahead) a ride in advance of their trip. Reservations give platforms precise information that describes the start time and locatio…
View article: Book-Ahead & Supply Management for Ridesourcing Platforms
Book-Ahead & Supply Management for Ridesourcing Platforms Open
Ridesourcing platforms recently introduced the ``schedule a ride'' service where passengers may reserve (book-ahead) a ride in advance of their trip. Reservations give platforms precise information that describes the start time and locatio…
View article: Performance Analysis of RSU-based Multihomed Multilane Vehicular Networks
Performance Analysis of RSU-based Multihomed Multilane Vehicular Networks Open
Motivated by the potentially high downlink traffic demands of commuters in future autonomous vehicles, we study a network architecture where vehicles use Vehicle-to-Vehicle (V2V) links to form relay network clusters, which in turn use Vehi…
View article: Performance Analysis of RSU-based Multihomed Multilane Vehicular\n Networks
Performance Analysis of RSU-based Multihomed Multilane Vehicular\n Networks Open
Motivated by the potentially high downlink traffic demands of commuters in\nfuture autonomous vehicles, we study a network architecture where vehicles use\nVehicle-to-Vehicle (V2V) links to form relay network clusters, which in turn\nuse V…
View article: Constrained Network Slicing Games: Achieving service guarantees and network efficiency
Constrained Network Slicing Games: Achieving service guarantees and network efficiency Open
Network slicing is a key capability for next generation mobile networks. It enables one to cost effectively customize logical networks over a shared infrastructure. A critical component of network slicing is resource allocation, which need…
View article: Resource Allocation for Network Slicing in Mobile Networks
Resource Allocation for Network Slicing in Mobile Networks Open
This paper provides a survey of resource allocation for network slicing. We focus on two classes of existing solutions: (i) reservation-based approaches, which allocate resources on a reservation basis, and (ii) share-based approaches, whi…
View article: Progressive Stochastic Greedy Sparse Reconstruction and Support Selection
Progressive Stochastic Greedy Sparse Reconstruction and Support Selection Open
Sparse reconstruction and sparse support selection, i.e., the tasks of inferring an arbitrary $m$-dimensional sparse vector $\mathbf{x}$ having $k$ nonzero entries from $n$ measurements of linear combinations of its components, are often e…
View article: Stochastic-Greedy++: Closing the Optimality Gap in Exact Weak Submodular Maximization
Stochastic-Greedy++: Closing the Optimality Gap in Exact Weak Submodular Maximization Open
Many problems in discrete optimization can be formulated as the task of maximizing a monotone and weak submodular function subject to a cardinality constraint. For such problems, a simple greedy algorithm is guaranteed to find a solution w…
View article: Performance-Complexity Tradeoffs in Greedy Weak Submodular Maximization with Random Sampling
Performance-Complexity Tradeoffs in Greedy Weak Submodular Maximization with Random Sampling Open
Many problems in signal processing and machine learning can be formalized as weak submodular optimization tasks. For such problems, a simple greedy algorithm (\textsc{Greedy}) is guaranteed to find a solution achieving the objective with a…
View article: Performance-Complexity Tradeoffs in Greedy Weak Submodular Maximization\n with Random Sampling
Performance-Complexity Tradeoffs in Greedy Weak Submodular Maximization\n with Random Sampling Open
Many problems in signal processing and machine learning can be formalized as\nweak submodular optimization tasks. For such problems, a simple greedy\nalgorithm (\\textsc{Greedy}) is guaranteed to find a solution achieving the\nobjective wi…
View article: Analysis of Data Harvesting by Unmanned Aerial Vehicles
Analysis of Data Harvesting by Unmanned Aerial Vehicles Open
International audience
View article: Modeling and Optimization of Human-machine Interaction Processes via the Maximum Entropy Principle
Modeling and Optimization of Human-machine Interaction Processes via the Maximum Entropy Principle Open
We propose a data-driven framework to enable the modeling and optimization of human-machine interaction processes, e.g., systems aimed at assisting humans in decision-making or learning, work-load allocation, and interactive advertising. T…
View article: Modeling and Analysis of Data Harvesting Architecture based on Unmanned\n Aerial Vehicles
Modeling and Analysis of Data Harvesting Architecture based on Unmanned\n Aerial Vehicles Open
This paper explores an emerging wireless Internet-of-things (IoT)\narchitecture based on unmanned aerial vehicles (UAVs). We consider a network\nwhere a fleet of UAVs at a fixed altitude flies on planned trajectories and IoT\ndevices on th…
View article: Modeling and Analysis of Data Harvesting Architecture based on Unmanned Aerial Vehicles
Modeling and Analysis of Data Harvesting Architecture based on Unmanned Aerial Vehicles Open
This paper explores an emerging wireless Internet-of-things (IoT) architecture based on unmanned aerial vehicles (UAVs). We consider a network where a fleet of UAVs at a fixed altitude flies on planned trajectories and IoT devices on the g…
View article: Network Slicing Games: Enabling Customization in Multi-Tenant Mobile Networks
Network Slicing Games: Enabling Customization in Multi-Tenant Mobile Networks Open
Network slicing to enable resource sharing among multiple tenants-network operators and/or services-is considered as a key functionality for next generation mobile networks. This paper provides an analysis of a well-known model for resourc…