Aaron Fainman

2 accepted papers

2026

GRAPH-AWARE LEARNING RATES FOR DECENTRALIZED OPTIMIZATION

ICASSP 2026poster

We propose an adaptive step-size rule for decentralized optimization. Choosing a step-size that balances convergence and stability is challenging. This is amplified in the decentralized setting as agents observe only local (possibly stochastic) gradients and global information (like smoothness) is u…

Cited by 0 Source ↗PDF ↗Scholar ↗

2025

Deep-Relative-Trust-Based Diffusion for Decentralized Deep Learning

Muyun Li, Aaron Fainman, Stefan Vlaski

ICASSP 2025accepted

Decentralized learning strategies allow a collection of agents to learn efficiently from local data sets without the need for central aggregation or orchestration. Current decentralized learning paradigms typically rely on an averaging mechanism to encourage agreement in the parameter space. We argu…

Cited by 0 Source ↗Scholar ↗