| layout | title | slug |
|---|---|---|
default |
Schedule |
schedule |
| Meeting | Notes | Topics | Homework (out) | Homework (due) | Readings** |
| Feb 23 | Decision-theoretic notation. Bayes estimator. Parametric families. Example: A/B testing. | 1.1, 1.2 | |||
| Feb 25 | Intrinsic loss functions. Examples of derivation of Bayes estimators. Doob's consistency theorem. | 2.1, 2.3, 2.5, 3.3 | |||
| Mar 2 | Bring laptop in class* | Rapid model development with probabilistic programming (i.e. languages such as BUGS, JAGS, Stan, etc). Example: challenger data. | Assignment 1 | ||
| Mar 4 | Bring laptop in class* | Hierarchies and mixtures. Examples: launchers data, geyser data. | 10.1, 10.2 | ||
| Mar 9 | Model selection and averaging. Example: rabbit holes. | 7.1 | |||
| Mar 11 | Bring laptop in class* | Model selection and averaging, continued. | 7.3 | ||
| Mar 13 | Assignment 1 | ||||
| Mar 16 | Nested sampling, IS, SIS. Example: HMMs. | Project proposal | 6.1, 6.2 | ||
| Mar 18 | SMC, SMC samplers. Some theoretical properties. Example:tree inference. | ||||
| Mar 23 | Particle MCMC. Proof of correctness. Particle genealogies. | Project proposal | |||
| Mar 25 | Reversible jump MCMC. Example: multiplicative proposals. | Assignment 2 | 6.3 | ||
| Mar 30 | Reversible jump MCMC vs. Bayesian non-parametric methods. | ||||
| Apr 1 | Bayesian non-parametrics continued. | ||||
| Apr 6 | University closed. | ||||
| Apr 8 | Lab 1 | Sampling in non-conjugate BNP models, hierarchical models and the sequence memoizer. Example: language modelling. | Assignment 3 (optional) | ||
| Apr 13 | Lab 2 | Second lab: 5pm, computer room on main floor of ESB | |||
| Apr 16 | Assignment 2 | ||||
| Apr 20 | Assignment 3 (optional) | ||||
| Apr 25 | Project deadline |
\*If you do not have a laptop, you can alternatively form a team with someone with a laptop and work in team. If you have difficulties forming such a team, contact me.
\*\*Unless noted otherwise, readings are in 'The Bayesian Choice', C.P. Robert.