EasyDeepLearn

Write the vanilla SGD update rule and explain each term.

easy

Answer

  • θt+1  =  θt    η    gt{\theta}_{t} + 1\; = \;{\theta}_{t}\; - \;{\eta}\; \cdot \;g_{t}, where θ is the parameter vector, η the learning rate, and gtg_{t} = ∇_θ L(θt  batcht)L({\theta}_{t}\;\mathrm{batch}_{t}) is the stochastic gradient on the current mini-batch.
  • Each step is a noisy estimate of the true gradient — the noise itself acts as regularizer and helps escape saddle points.
  • Convergence rate for convex problems: O(1/√T).
Check yourself — multiple choice
  • θ ← θ + η · g (add gradient)
  • θ ← θ - η · g — noisy estimate on mini-batch, regularizes and escapes saddles
  • SGD uses second-order info
  • SGD requires full-batch gradient

Vanilla SGD: subtract learning rate × mini-batch gradient.

#optimizers#optimization

Practise Deep Learning

214 interview questions in this topic.

Related questions