Skip to content
ml-recall

#linear-layer

2 questions from Optimization & Calculus, Text Representation & Embeddings

Timed test

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

Take a test
Q 1 / 2Score 0 / 0

Question 1: Single choice

hardAI-checkedOptimization & Calculus

Let W∈Rn×mW \in \mathbb{R}^{n \times m} have rows w1⊤,…,wn⊤\mathbf{w}_1^\top, \ldots, \mathbf{w}_n^\top, let v∈Rm\mathbf{v} \in \mathbb{R}^m, and let f:R→Rf: \mathbb{R} \to \mathbb{R} be differentiable. What is the gradient of J(W)=∑k=1nf(wk⊤v)J(W) = \sum_{k=1}^n f(\mathbf{w}_k^\top \mathbf{v}) with respect to WW (an n×mn \times m matrix)? Here f′f' acts element-wise on the vector WvW\mathbf{v}.

Choose one option