Nonparametric Bayesian Methods: Models, Algorithms, and Applications I

on YouTube ↗01h 06m

" summary

Argument: treat the latent mixture as an infinite parameter so realized cluster count grows with data (roughly log N in the basic Dirichlet case). Construction: use Beta/Dirichlet stick‑breaking (the stick‑breaking GEM), or the Chinese‑restaurant marginalization, and sample with Gibbs/HMC or truncate the stick. Practical point: only finitely many atoms appear in finite data, so on‑demand generation and truncation let simulation and posterior inference work efficiently in practice.

AI-written summary · about

" chapters (6)

1. 0:00Introduction to the topic
2. 3:34Defining nonparametric Bayes
3. 11:02Finite mixture models
4. 26:22Challenges of model size
5. 42:16Dirichlet processes and GEM
6. 59:00Inference and conclusion

" related talks

• 48mLocally Private Bayesian Inference for Count ModelsSteven Wu · Privacy and the Science of Data Analysis 2019
• 43mIdentifying Mixtures Of Bayesian Network DistributionsYuval Rabani · Algorithmic Aspects of Causal Inference 2022
• 42mBayesian Non-parametric Models for Data Science using PyMC3Christopher Fonnesbeck · PyCon 2018
• 18mConstantineConstantine · StanCon 2018
• 25mScaling Bayesian ExperimentationRobert J. Neal, Paul Cho · Scale By The Bay 2018
• 32mNew Approaches To Learning Nonparametric (Latent) Causal Graphical ModelsBryon Aragam · Algorithmic Aspects of Causal Inference 2022
• 01h 06mTutorial: Methods from Statistical Physics IIAhmed El Alaoui · Deep Learning Theory Workshop and Summer School 2022
• 44mMixture Models, Robustness, and Sum of Squares ProofsJerry Li · Robust and High-Dimensional Statistics (Simons Institute workshop, 2018)