Multitask Online Learning: Listen to the Neighborhood Buzz
This work addresses decentralized learning for agents in networked environments, offering incremental improvements in regret bounds based on task similarities.
The paper tackles multitask online learning in decentralized networks where agents communicate only with neighbors, introducing the MT-CO2OL algorithm whose regret improves with task similarity and network structure, and it can be made differentially private with minimal regret impact, supported by experiments.
We study multitask online learning in a setting where agents can only exchange information with their neighbors on an arbitrary communication network. We introduce $\texttt{MT-CO}_2\texttt{OL}$, a decentralized algorithm for this setting whose regret depends on the interplay between the task similarities and the network structure. Our analysis shows that the regret of $\texttt{MT-CO}_2\texttt{OL}$ is never worse (up to constants) than the bound obtained when agents do not share information. On the other hand, our bounds significantly improve when neighboring agents operate on similar tasks. In addition, we prove that our algorithm can be made differentially private with a negligible impact on the regret. Finally, we provide experimental support for our theory.