Thompson Sampling for Machine Learning

on YouTube ↗35m

" summary

Overview of applying Thompson Sampling to machine learning problems, including technical examples like using Bayesian Neural Networks implemented with Edward. Discusses recent developments and encourages thinking strategically about how models should learn from new incoming data.

AI-written summary · about

" chapters (12)

1. 0:00Thompson Sampling intro
2. 1:29Multi-arm bandit
3. 3:33Thompson Sampling uses
4. 6:25Exploration vs exploitation
5. 9:56How sampling works
6. 13:30Thompson Sampling trials
7. 17:00Prior distributions role
8. 20:30Interaction effects capture
9. 24:23Research and medical apps
10. 27:11Testing Thompson Sampling
11. 30:35Setting up experiments
12. 33:11Key takeaways

" related talks

• 41mThompson sampling for personalising a car brands advertisementsN. van Engelenhoven, J. Hamerlinck · PyData Eindhoven 2022
• 39mSaaaS - Sampling as an Algorithm ServiceVincent D Warmerdam · PyData London 2017
• 48mAdaptivity and Confounding in Multi-armed Bandit ExperimentsDaniel Russo · Quantifying Uncertainty: Stochastic, Adversarial, and Beyond
• 32mDesigning Adaptive Experiments For Policy Learning And InferenceMolly Offer-Westort · Learning from Interventions 2022
• 17mByron Galbraith, Chief Data Scientist, Talla, NYC 2017Byron Galbraith · MLconf NYC 2017
• 01h 09mStochastic Bandits: Foundations and Current PerspectivesShipra Agrawal · Data-Driven Decision Processes Boot Camp 2022
• 31mInterpreting Machine Learning From the Perspective of Nonequilibrium SystemsDavid Limmer · Interpretable Machine Learning in Natural and Social Sciences (2022)
• 30mMachine Learning Techniques for Class Imbalances & AdversariesBrendan Herger · PyData DC 2016