Model techniques map
Techniquesoptimizationoptimizer

implementation detail · filed under optimization

Nesterov momentum

Applying Nesterov momentum in the Muon update or orthogonalization process.

Also called Nesterov-accelerated momentum, Nesterov trick.

sources
2
models
3
labs adopt it
2
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 2

Documented in

Evidence

2 spans quoted from the sources, strongest treatment first.

we also apply weight decay to Muon parameters, use the Nesterov trick

usedoptimizationin DeepSeek-V4DeepSeek

The NS iteration is applied to a Nesterov-accelerated momentum

usedoptimizationin Qwen3.8-NextQwen

Filed alongside

Other methods under optimization :: optimizer.