Living a few months in Cambridge: (red is day, blue is night)

I love the blue mass at the astronomy centre and the movie theatre. More random projects here.

An interesting statement: “neural networks use finitely many highly adaptive basis functions whereas gaussian processes typically use infinitely many fixed basis functions” - paraphrased from Wilson et al. 2015, based on work by MacKay, Neal and others.

Models seem to be constructed by making them likely in the light of data but not always such that they’re able to generate something similar. More ramblings about the likelihood.

# Recently Updated:

### An Ising-Like Model

## … using Stan & HMC

### Sparse Gaussian Process Examples

## A Minimal Working Example

### Random Stuff

## Random Stuff

### Stochastic Bernoulli Probabilities

Consider: