Machine Learning for Game Developers
summary
Overview of practical ML for games with concrete tools and results: cloud APIs like Translate, Vision, Cloud Natural Language and Video Intelligence are immediate REST endpoints for chat translation, object/landmark detection and sentiment scoring; AutoML Vision trains on labeled images (example dataset ~1,800 images, 12 hour training → ~90% accuracy, confusion matrix shows classes with 500 images ~90% and smaller classes ~60%); AutoML Tables handles structured data; TensorFlow models train on Compute Engine or use TPUs for huge speedups (example 24h → 15min); pipeline uses BigQuery and Cloud Storage for logs and trained models, enabling QA bots, smarter NPCs and fraud/cheat detection.
AI-written summary · about
chapters (5)
related talks
47mBuilding smarter apps with Machine Learning, from magic to reality
42mMachine Learning APIs, from Training [...]
54mBuilding smarter apps with Machine Learning, from magic to reality
59mMachine Intelligence at Google Scale
57mBuilding smarter solutions with Machine Learning, from magic to reality
14mMachine Learning for developers in a hurry
47mBuilding smarter solutions without machine learning expertise
42mUse Machine Learning in your code, without being a ML expert