Machine learning algorithms, choosing the correct algorithm for your problem
summary
Clear rules for matching problems to supervised vs unsupervised vs semi‑supervised vs reinforcement learning, with concrete examples: predict car price or spam using decision trees for labeled data, find customer segments with k‑means on unlabeled data, auto‑label large image archives using mixed strategies, and train agents by reward where AlphaGo and OpenAI Dota 2 show reinforcement learning solving tasks people thought decades away. Explains features vs labels and when data quantity or labeling cost forces choice of algorithm.
AI-written summary · about
chapters (5)
- 0:00Start
- 3:03Decision trees
- 3:53Unsupervised learning
- 5:34Semi-supervised learning
- 6:22Reinforcement learning
related talks
26mHow Machine Learning Can Help
44mHow to Improve UX with Machine Learning: Insights and learnings until now
51mMachine Learning Exposed Introduction to Machine Learning
32mLearning from algorithms to make Human decisions
41mBeyond the Algorithm: What Makes Machine Learning Work?
27mData Science and Neural Networks - Big Love or just a Love Affair?
45mAGT and Machine Learning
43mDebugging AI