Marano, StefanoSayed, Ali H.2021-03-262021-03-262021-03-262020-01-0110.1109/ICASSP40776.2020.9053560https://infoscience.epfl.ch/handle/20.500.14299/176380WOS:000615970405157Adaptation and learning over multi-agent networks is a topic of great relevance with important implications. Elaborating on previous works on single-task networks engaged in decision problems, here we consider the multi-task version in the challenging scenario where the state of nature may change arbitrarily. We propose a data diffusion scheme for tracking these changes in real time, and investigate by numerical simulations the corresponding steady-state decision performance. For the slow-adaptation regime, the complete analytical characterization of the agents' status is provided, under the simplifying assumption that the network connection matrix is correctly estimated.AcousticsEngineering, Electrical & ElectronicEngineeringadaptive networksdiffusion schemesmulti-task decisionsslow-adaptation regimeparameter estimationconsensusalgorithmstext::conference output::conference proceedings::conference paper