Skip to content
ml-recall

#nesterov

2 questions from Optimization & Calculus

Timed test

2 questions · 2 min · score and explanations at the end

Take a test
Q 1 / 2Score 0 / 0

Question 1: Short answer

mediumAI-checkedOptimization & Calculus

Nesterov momentum in moving-average form uses vt=βvt−1+(1−β) ∇L(θ−ηβvt−1)v_t = \beta v_{t-1} + (1 - \beta)\,\nabla L(\theta - \eta\beta v_{t-1}) and θ←θ−ηvt\theta \leftarrow \theta - \eta v_t. For L(θ)=θ2L(\theta) = \theta^2 with θ=1\theta = 1, vt−1=1v_{t-1} = 1, β=0.9\beta = 0.9 and η=0.1\eta = 0.1, the gradient used in this step is and the new θ\theta is . Give 4 decimals.

Fill in every blank (hints in the boxes). Enter moves to the next blank.