Decentralized Parameter-Free Online Learning
This addresses the need for robust multi-agent learning in distributed sensing and collaborative ML, though it appears incremental as it extends existing coin-betting methods to decentralized settings.
The paper tackles the problem of decentralized online learning without hyperparameter tuning by proposing the first parameter-free algorithms with network regret guarantees, achieving sublinear regret validated through experiments on synthetic and real datasets.
We propose the first parameter-free decentralized online learning algorithms with network regret guarantees, which achieve sublinear regret without requiring hyperparameter tuning. This family of algorithms connects multi-agent coin-betting and decentralized online learning via gossip steps. To enable our decentralized analysis, we introduce a novel "betting function" formulation for coin-betting that simplifies the multi-agent regret analysis. Our analysis shows sublinear network regret bounds and is validated through experiments on synthetic and real datasets. This family of algorithms is applicable to distributed sensing, decentralized optimization, and collaborative ML applications.