Batch Normalization Intuition - Quant Trader Interview Question
Difficulty: Medium
Category: Linear Algebra & Machine Learning
Practice quant interview questions from top firms including Jane Street, Citadel, Two Sigma, DE Shaw, and other leading quantitative finance companies.
Topics: machine-learning, deep-learning, batch-normalization, neural-networks
Problem Description
You're training a deep neural network for image classification. You observe that the training process is unstable, and the network learns very slowly. Considering techniques to improve training, what is the primary problem that batch normalization is designed to address?
Practice this medium trader interview question on MyntBit - the all-in-one quant learning platform with 200+ quant interview questions for Jane Street, Citadel, Two Sigma, and other top quantitative finance firms.