PracHub
QuestionsPremiumCoachesLearningGuidesInterview Prep
|Home/Machine Learning/Snapchat

Explain Core ML Concepts

Last updated: Apr 22, 2026

Quick Overview

This question evaluates understanding of foundational machine learning and deep learning concepts, including the bias–variance decomposition, differences between batch normalization and layer normalization, and the causes of vanishing gradients.

  • medium
  • Snapchat
  • Machine Learning
  • Machine Learning Engineer

Explain Core ML Concepts

Company: Snapchat

Role: Machine Learning Engineer

Category: Machine Learning

Difficulty: medium

Interview Round: Onsite

You are interviewing for a machine learning role. Answer the following core questions: 1. Explain the bias-variance decomposition of prediction error and how it relates to underfitting and overfitting. 2. Compare batch normalization and layer normalization: how each is computed, where each works well, and their trade-offs. 3. Explain why vanishing gradients happen in deep neural networks and describe practical ways to mitigate them.

Quick Answer: This question evaluates understanding of foundational machine learning and deep learning concepts, including the bias–variance decomposition, differences between batch normalization and layer normalization, and the causes of vanishing gradients.

Related Interview Questions

  • Explain Overfitting and Transformer Attention - Snapchat (medium)
  • Discuss ML Project Tradeoffs - Snapchat (medium)
  • Model an ads ranking system - Snapchat (medium)
  • Explain BatchNorm, optimizers, and L1/L2 - Snapchat (medium)
  • Explain CLIP, contrastive losses, and retrieval limits - Snapchat (medium)
Snapchat logo
Snapchat
Feb 2, 2026, 12:00 AM
Machine Learning Engineer
Onsite
Machine Learning
2
0
Loading...

You are interviewing for a machine learning role. Answer the following core questions:

  1. Explain the bias-variance decomposition of prediction error and how it relates to underfitting and overfitting.
  2. Compare batch normalization and layer normalization: how each is computed, where each works well, and their trade-offs.
  3. Explain why vanishing gradients happen in deep neural networks and describe practical ways to mitigate them.

Solution

Show

Submit Your Answer to Earn 20XP

Sign in to leave a comment

Loading comments...

Browse More Questions

More Machine Learning•More Snapchat•More Machine Learning Engineer•Snapchat Machine Learning Engineer•Snapchat Machine Learning•Machine Learning Engineer Machine Learning
PracHub

Master your tech interviews with 8,000+ real questions from top companies.

Product

  • Questions
  • Learning Tracks
  • Interview Guides
  • Resources
  • Premium
  • For Universities
  • Student Access

Browse

  • By Company
  • By Role
  • By Category
  • Topic Hubs
  • SQL Questions
  • Compare Platforms
  • Discord Community

Support

  • support@prachub.com
  • (916) 541-4762

Legal

  • Privacy Policy
  • Terms of Service
  • About Us

© 2026 PracHub. All rights reserved.