Arindam Banerjee

Sketched Adaptive Federated Deep Learning: A Sharp Convergence Analysis

2024 · arXiv (Cornell University)

Combining gradient compression methods (e.g., CountSketch, quantization) and adaptive optimizers (e.g., Adam, AMSGrad) is a desirable goal in federated learning (FL), with potential benefits on both fewer communication rounds and less per-round communication. In spite …

Optimization for Neural Operators can Benefit from Width

2025 · arXiv (Cornell University)

Neural Operators that directly learn mappings between function spaces, such as Deep Operator Networks (DONs) and Fourier Neural Operators (FNOs), have received considerable attention. Despite the universal approximation guarantees for DONs and FNOs, there is …