Skip to content
ml-recall

#prelu

3 questions from Neural Network Basics

Timed test

3 questions · 3 min · score and explanations at the end

Take a test
Q 1 / 3Score 0 / 0

Question 1: Short answer

easyAI-checkedNeural Network Basics

The parametrized ReLU is pReLU(x)=max⁡(0,x)+αmin⁡(0,x)\mathrm{pReLU}(x) = \max(0, x) + \alpha\min(0, x). With α=0.25\alpha = 0.25, a unit has pre-activation x=−3x = -3 and receives the upstream gradient ∂L/∂ pReLU(x)=2\partial L / \partial\,\mathrm{pReLU}(x) = 2. What is ∂L/∂x\partial L / \partial x?

a number