Probabilistic Programming: What It Is and How It Works

on YouTube ↗29m

" summary

Builds a simple Scala model that explains seeing cats by a hidden enticement with probabilities milkshake 0.6 fish 0.1 nothing 0.3, shows exact enumeration for small discrete cases and sampling for larger/continuous ones, and implements a distribution as a Free monad with primitive generators and an interpreter that turns a Distribution into a RNG -> A sampler (uses java.util.Random, SBT demo). Posterior calculation for three observed cats yields posterior milkshake ≈ 85%. Code exists but is noted as rough.

AI-written summary · about

" chapters (12)

1. 0:00Introduction
2. 1:53Probability Distribution
3. 2:57Conditional Probability Distribution
4. 3:56Portability
5. 4:15Independent Events
6. 5:05Example Query
7. 5:53Example Discrete
8. 10:17Approximate Solutions
9. 10:47Sampling
10. 12:16Tailoring
11. 13:33Free Modern
12. 20:49Deep Learning vs probabilistic programming

" related talks

• 42mBayes and Bool Walk into a BarSamy Zarour · Scala.IO 2017
• 29mMotivating Probabilistic ProgrammingRahul Chitturi · Scale By The Bay 2018
• 31mProbabilistic Programming: The What, Why and HowMaria I. Gorinova · SPLASH 2020 REBASE
• 30mProbabilistic modelling with ScalaMartin Mauch · Scalar Conference 2015
• 01h 03mA Personal Viewpoint on Probabilistic ProgrammingDaniel Roy · Uncertainty in Computation 2016
• 43mProbabilistic Programming in the Real WorldZach Anglin · PyData DC 2018
• 36mProbabilistic Programs Which Make (Common) SenseZenna Tavares · Strange Loop 2015
• 43mThe Power of Probabilistic ProgrammingBen Vigoda · O'Reilly Artificial Intelligence Conference 2017 - New York, New York