Status Updates From Pattern Recognition and Mac...

hermoso

n+1 attempts reading this textbook....

En pause

Heavy

Why do I just now understand the connection between expectation maximization and k-means?

Golden!

I've been meaning to read this all-time classic for a while. Need to finally get this done cover-to-cover.

More interesting than I thought. Bishop has done a great introduction of probability, decision and information theories which are amazingly elaborated for the sake of the rest of the book.

Textbook for information theory and machine learning courses.

Not gonna lie , I gave up on chapter 2 and just read the slides here instead http://lear.inrialpes.fr/~jegou/bisho... which has the cliff notes and useful graphs from the textbook. Good overview of parametric vs nonparameteric methods for density estimation. Intro to common probability distribution functions. My favorite concepts are Gaussian mixture models and Kernal density estimators.

2.3 is dense! I'm not sure if I want to study the entire derivation of Gaussian distribution. The first 8 minutes of this video on Gaussian distribution seems to cover the important properties https://www.youtube.com/watch?v=50Vgw... Will attempt to read it after 2.4

The little biographical vignettes are very nice. I'd never heard about the feud between Jacob Bernoulli and his brother...

It is said that von Neumann recommended to Shannon that he use the word entropy, not only because of its similarity to the quantity used in physics but also because "nobody knows what entropy really is, so in any discussion you will always have an advantage."

Nearly all the mass of a high-dimensional sphere is concentrated in a thin shell near the surface. So be careful about analogising from two dimensions to many.

If this book were a spaceship, we'd now be out of the solar system and about to engage the hyperdrive. Fasten your seat belts.

My knowledge of statistics, I will claim, is not a hazy 'un/And I know why you should and why you shouldn't be a Bayesian. The Major-General

Starts off well, with a brilliant example illustrating the key concepts of data sparseness and overfitting.

Just arrived!