4 problems

Consider decentralized learning on a directed graph in which communication edge weights are optimized using the agents' data distributions. An optimized topology is the resulting w…

In directed distributed learning, let the communication graph topology be refined dynamically by adjusting its edge weights during training. Dynamic topology refinement conjecture.…

Let $f_i$ denote the local loss function associated with client $i$. Consider decentralized learning settings in which the algorithms either do not suffer from, or actively elimina…

In decentralized stochastic optimization, let gradient and Hessian estimators be constructed using mini-batches in the proposed variance-reduced stochastic second-order methods. La…