Mikail Khona

Mikail Khona is a research scientist at NVIDIA. He joined the NVIDIA Applied Deep Learning Research group in 2024 after finishing his PhD at the Massachusetts Institute of Technology. Mikail’s technical focus has been pretraining, optimization, and model architecture.
Avatar photo

Posts by Mikail Khona

Developer Tools & Techniques

Advancing Emerging Optimizers for Accelerated LLM Training with NVIDIA Megatron

Higher-order optimization algorithms such as Shampoo have been effectively applied in neural network training for at least a decade. These methods have achieved... 9 MIN READ