arXiv subjects

Idoia de la Iglesia

Publications and source records attributed to Idoia de la Iglesia.

3 recordsLinked to original sources

arXiv2026

A Comparative Study of CNN Optimization Methods for Edge AI: Exploring the Role of Early Exits

Deploying deep neural networks on edge devices requires balancing accuracy, latency, and resource constraints under realistic execution conditions. To fit models within these constraints, two broad strategies have emerged: static compression techniques such as pruning and quantization, which permanently reduce model size, and dynamic approaches such as early-exit mechanisms, which adapt computational cost at runtime. While both families are widely studied in isolation, they are rarely compared under identical conditions on physical hardware. This paper presents a unified deployment-oriented comparison of static compression and dynamic early-exit mechanisms, evaluated on real edge devices using ONNX based inference pipelines. Our results show that static and dynamic techniques offer fundamentally different trade-offs for edge deployment. While pruning and quantization deliver consistent memory footprint reduction, early-exit mechanisms enable input-adaptive computation savings that static methods cannot match. Their combination proves highly effective, simultaneously reducing inference latency and memory usage with minimal accuracy loss, expanding what is achievable at the edge.

cs.AI↗

arXiv2026

On-Device Deep Reinforcement Learning for Decentralized Task Offloading Performance trade-offs in the training process

Allowing less capable devices to offload computational tasks to more powerful devices or servers enables the development of new applications that may not run correctly on the device itself. Deciding where and why to run each of those applications is a complex task. Therefore, different approaches have been adopted to make offloading decisions. In this work, we propose a decentralized Deep Reinforcement Learning (DRL) agent to address the selection of computing locations. Unlike most existing work, we analyze it in a real testbed composed of various edge devices running the agent to determine where to execute each task. These devices are connected to a Multi-Access Edge Computing (MEC) server and a Cloud server through 5G communications. We evaluate not only the agent's performance in meeting task requirements but also the implications of running this type of agent locally, assessing the trade-offs of training locally versus remotely in terms of latency and energy consumption.

cs.ET↗

arXiv2014

NAVI: Neighbor Aware Virtual Infrastructure for Information Dissemination in Vehicular Networks

Vehicular Networks enable a vast number of innovative applications, which rely on the efficient exchange of information between vehicles. However, efficient and reliable data dissemination is a particularly challenging task in the context of vehicular networks due to the underlying properties of these networks, limited availability of network infrastructure and variable penetration rates for distinct communication technologies. This paper presents a novel system and mechanism for information dissemination based on virtual infrastructure selection in combination with multiple communication technologies. The system has been evaluated using a simulation framework, involving network simulation in conjugation with realistic vehicular mobility traces. The presented simulation results show the feasibility of the proposed mechanism to achieve maximum message penetration with reduced overhead. Compared with a cellular-based only solution, our mechanism shows that the judicious vehicle selection can lead to improved network utilization through the offload of traffic to the short-range communication network.

cs.NI↗