Conceptexpected objective-gap bound

Expected Objective Gap Bound

A quantity in optimization theory that characterizes convergence rates, representing the expected gap in objective function value between the current parameters and the optimal solution. It measures how far the expected objective value remains from the optimum and is used to bound the rate at which an optimization algorithm approaches the optimal solution.

Timeline (last 90 days)

Nov 2

Under explicit regularity and alignment assumptions, the Muon-RT paper establishes a convergence result in which the expected objective gap bound contracts linearly toward the residual error level.

Unverified50%

All Facts (1)

Unverified

Under explicit regularity and alignment assumptions, the Muon-RT paper establishes a convergence result in which the expected objective gap bound contracts linearly toward the residual error level.

50%

Source Articles

Muon-RT: Reusing Teacher Trajectories for Momentum-Free LLM Self-DistillationNov 2