Bayesian Decision Analysis [Tutorial]
summary
Hands-on walkthrough of Bayesian decision analysis with runnable Jupyter notebooks (Colab), using NumPy, SciPy and Pandas to represent PMFs and implement updates. Covers Bayes’ theorem to Thompson sampling for multi-armed bandits, showing the actual code: pmf as a pandas Series, pmf_choice via numpy.random.choice, flip/play simulation and a choose–play–update loop. Demonstration shows that after ~100 trials the algorithm concentrates on the best arm (often ~50–60% of plays). Also sketches optimal bidding and decision‑rule examples and supplies solution notebooks.
AI-written summary · about
chapters (15)
- 0:00Bayesian Decision Analysis intro
- 4:55Speaker background
- 6:43Bayes theorem basics
- 13:26Five urn problem solved
- 21:38Jupyter notebook setup
- 25:57First break recap
- 26:17Implementing five urns
- 30:54From urns to bandits
- 39:47Updating beliefs
- 52:54Second break recap
- 57:56Simulating machines
- 1:04:00Thompson sampling
- 1:16:17Notebook review & Q&A
- 1:18:07Summary & examples
- 1:22:29Q&A
related talks
21mLeveraging Multi-Armed Bandit Algorithms for Dynamic Decision Making
39mThe Duct Tape of Heroes: Bayes Rule
01h 27mProbabilistic Python: An Introduction to Bayesian Modeling with PyMC
01h 27mBayesian Time Series
01h 31mProbabilistic Programming and Bayesian Inference in Python
30mPractical Multi Armed Bandits
42mSolving Real-World Business Problems with Bayesian Modeling
01h 26mProbabilistic Programming and Bayesian Computing with PyMC