Blog of a data person.
LLM tricks that I *do* find interesting.
For the Office, and for the Horde!
Revisting the Google Emotions dataset.
What lies in "Research Advocacy"?
GridSearch is Not Enough: Part Seven
Horrible Remedies for Broken Recommenders
GridSearch is Not Enough: Part Six
Just Another Dangerous Situation
Let's make life a whole log simpler.
GridSearch is Not Enough: Part Five
Making impact without touching the core of a library.
GridSearch is Not Enough: Part Four
Put Cookie-Monster on a Diet.
Reduce Common Sense to Calculus.
Some assumptions are a bit weird.
We should evolve beyond it.
GridSearch is not enough: Part Three.
An Ode to Pipes, Seeds and Simplicity
GridSearch is not Enough: Part Two.
GridSearch is not Enough: Part One.
At best, it's whats actually being predicted.
Sort of came up with an alternative to VI here.
Combine the Neural with the Normal.
How to search in search space.
A simple introduction to PyMC3.
A convenient overlap.
Stick a Network in the Q-learning algorithm.
Join me in preventing this.
Merely *a* argument, but one that I like.
Some observations but also downsides.
How I got an A+ for measuring the wrong thing.