Yuge Shi

Yuge Shi

DPhil student, University of Oxford

Posts by Tags

Bayesian

Gaussian Process, not quite for dummies

19 minute read

Published:

Before diving in

For a long time, I recall having this vague impression about Gaussian Processes (GPs) being able to magically define probability distributions over sets of functions, yet I procrastinated reading up about them for many many moons. However, as always, I’d like to think that this is not just due to my procrastination superpowers. Whenever I look up “Gaussian Process” on Google, I find these well-written tutorials with vivid plots that explain everything up until non-linear regression in detail, but shy away at the very first glimpse of any sort of information theory. The key takeaway is always,

A Gaussian process is a probability distribution over possible functions that fit a set of points.

Computer vision

ELBO

Gaussian Process

Gaussian Process, not quite for dummies

19 minute read

Published:

Before diving in

For a long time, I recall having this vague impression about Gaussian Processes (GPs) being able to magically define probability distributions over sets of functions, yet I procrastinated reading up about them for many many moons. However, as always, I’d like to think that this is not just due to my procrastination superpowers. Whenever I look up “Gaussian Process” on Google, I find these well-written tutorials with vivid plots that explain everything up until non-linear regression in detail, but shy away at the very first glimpse of any sort of information theory. The key takeaway is always,

A Gaussian process is a probability distribution over possible functions that fit a set of points.

Machine Learning

Gaussian Process, not quite for dummies

19 minute read

Published:

Before diving in

For a long time, I recall having this vague impression about Gaussian Processes (GPs) being able to magically define probability distributions over sets of functions, yet I procrastinated reading up about them for many many moons. However, as always, I’d like to think that this is not just due to my procrastination superpowers. Whenever I look up “Gaussian Process” on Google, I find these well-written tutorials with vivid plots that explain everything up until non-linear regression in detail, but shy away at the very first glimpse of any sort of information theory. The key takeaway is always,

A Gaussian process is a probability distribution over possible functions that fit a set of points.

Monte Carlo gradient estimation

Self-supervised Learning

Uncertainty

Gaussian Process, not quite for dummies

19 minute read

Published:

Before diving in

For a long time, I recall having this vague impression about Gaussian Processes (GPs) being able to magically define probability distributions over sets of functions, yet I procrastinated reading up about them for many many moons. However, as always, I’d like to think that this is not just due to my procrastination superpowers. Whenever I look up “Gaussian Process” on Google, I find these well-written tutorials with vivid plots that explain everything up until non-linear regression in detail, but shy away at the very first glimpse of any sort of information theory. The key takeaway is always,

A Gaussian process is a probability distribution over possible functions that fit a set of points.

Unsupervised Learning

VAE