Decentralised Collaborative Edge AI for Resource Sharing
summary
Claims a concrete architecture: peer-to-peer federated learning plus inter-agent knowledge distillation and distributed continual learning running on microcontrollers and TinyML. Devices such as UAVs robots embedded sensors and wearables offload tasks, share neurons and weights over low-power links. Demonstrations include disaster management farming and pollution monitoring; experiments cite resource pooling like 40 MB from ten wearables. Papers describe routing protocols and convergence behavior and report prototypes that operate under intermittent connectivity and tight energy budgets.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 2:34Edge AI to resource sharing
- 6:49How it works
- 10:42Tiny device challenges
- 14:34Future directions
related talks
48mFederated ML for Edge Applications
34mLFN AI Taskforce: Platform-Driven Fede...
12mFederated Learning në ML/AI - mësimi i decentralizuar
34mDecentralized Machine Learning
12mUsing AI-enabled Edge Cloud Technologies to Strengthen Europe’s...
42mThe Stakes of Green AI for Embedded Systems
31mInfiniEdge AI: Bringing Open-Source Generative AI To the Edge
21mA New Architecture for the Intelligent Edge
