Approximating Integrals with Importance Sampling

Short blurb on approximating integrals with importance sampling

Gaussian Noise and Mean Squared Error

Parameter estimation in linear regression from 2 perspectives.

Bayesian Hierarchical Classification in Numpyro

In this post I use Numpyro to build a Bayesian hierarchical model to classify Amazon products into a taxonomy.

Hierarchical Models in Numpyro

Exploring pooling and hierarchical models with Numpyro by estimating the free throw percentage of NBA players

Gaussian Processes and Regression

A explanation of Gaussian processes and Gaussian process regression, starting with simple intuition and building up to inference. I sample from a GP in native Python and test GPyTorch on a simple simulated example.

Dealing with Overconfidence in Neural Networks: Bayesian Approach

I trained a classifier on images of animals and gave it an image of myself, it's 98% confident I'm a dog. This is an exploration of a possible Bayesian fix. Code available too

Notes on the Beta and Dirichlet Distributions

The Beta and Dirichlet distributions are related to each other in a similar way to the Binomial and Multinomial distributions. This post explains the relationship between these 4 distributions using a simple example and some code.

Bayesian Changepoint Detection of COVID-19 Cases in Pyro

Used Pyro and a Bayesian changepoint model to detect the date that COVID-19 cases started to flattern in different countries.