no code implementations • 8 Feb 2023 • Pranay Sharma, Rohan Panda, Gauri Joshi
We analyze the convergence of the proposed algorithm for classes of nonconvex-concave and nonconvex-nonconcave functions and characterize the impact of heterogeneous client data, partial client participation, and heterogeneous local computations.
no code implementations • 9 Mar 2022 • Pranay Sharma, Rohan Panda, Gauri Joshi, Pramod K. Varshney
In this paper, we consider nonconvex minimax optimization, which is gaining prominence in many modern machine learning applications such as GANs.