Distributed Machine Learning in IoT and Vehicular Networks

In this talk we discuss approaches for distributed machine learning (ML) in wireless networks, in particular vehicular networks and resource-constrained Internet of Things (IoT) networks. Federated Learning (FL) and Split Learning (FL) are popular approaches in such environments.
First, we present Early Exit of Communication (EEoC), which adaptively splits ML inference in an IoT edge computing environment to meet latency and energy constraints. This layer-based (vertically partitioned) approach has been extended by Distributed Micro-Split Deep Learning in Heterogeneous Dynamic IoT (DISNET), which adds horizontal partitioning to better support flexible, distributed, and parallel execution of neural network models on heterogeneous IoT devices under dynamic conditions. Then, we also consider the training aspect by developing and evaluating Adaptive REsource-aware Split-learning (ARES), a scheme for efficient model training in IoT systems. Recent work suggests Dynamic FL (DFL) for heterogeneous IoT, which uses resource-aware SL and FL based on similarity-based layer-wise model aggregation.
Co-sponsored by: Universidad Tecnológica de Panamá
Speaker(s): Torsten Braun
Agenda:
Registro de participantes
Conferencia
Preguntas y respuestas
Despedida
Room: Salón Roberto Barraza, Bldg: Edificio 1, Salón Roberto Barraza, Universidad Tecnológica de Panamá, Campus Víctor Levi Sasso, Panama, Panama, Panama