Glossary

Limitations:

Behavior differs between training and evaluation modes (source of subtle bugs) - Performance degrades with very small batch sizes (batch statistics become noisy) - Not ideal for sequence models where batch statistics mix different sequence lengths

Learn More

AI Engineering — Chapter 12: Training Deep Networks

AI Engineering — Chapter 25: Alignment: RLHF, DPO, and Beyond

College Football Analytics — Chapter 2: The Data Landscape of NCAA Football

Data & Society — Chapter 6: Ethical Frameworks for the Data Age

Data & Society — Chapter 18: Generative AI: Ethics of Creation and Deception

Data & Society — Chapter 27: Data Stewardship and the Chief Data Officer

Data & Society — Chapter 31: Misinformation, Disinformation, and Platform Governance

How Humans Get Stuck — Chapter 34: Adversarial Collaboration and Other Tools for Producing Less Wrong Knowledge

How Your House Works — Chapter 17: Solar, EV Chargers, and Whole-Home Generators: Modern Electrical Additions

How Your House Works — Chapter 18: How Heating Systems Work: Furnaces, Boilers, and Heat Pumps

Prediction Markets — Chapter 14: Binary Outcome Trading Strategies

Basketball Analytics — Chapter 13: Win Shares and Wins Above Replacement

Basketball Analytics — Chapter 15: Player Tracking Analytics

Soccer Analytics — Chapter 2: Data Sources and Collection in Soccer

Soccer Analytics — Case Study 6.1: Mapping Barcelona's Tiki-Taka Through Spatial Analysis

Soccer Analytics — Case Study 6.2: Building a Defensive Shape Analyzer Using Coordinate Data

Soccer Analytics — Chapter 17: Spatial Analysis and Pitch Control

Propaganda — Chapter 29: Counter-Propaganda, Strategic Communication, and Prebunking

Science of Seduction — Case Study 19.1: Moore's (1985) Ethological Study of Women's Courtship Signals in Bars

Sports Betting — Chapter 2: Probability and Odds

Sports Betting — Chapter 26: Ratings and Ranking Systems

Related Terms