mossr / beautifulalgorithms.jl Goto Github PK

Hi and thank you for this package.
The algorithms are indeed very beautiful!

I find your value_iteration.jl especially useful.
Is there a chance you can include some other common algorithms for solving dynamic programming problems?
A recent paper provides 7 common algorithms to solve a routine dynamic programming problem:

(the AR process will be discretized into a discrete state Markov Chain)

Optimality conditions:

7 standard methods:

Hello,

I am trying to run the test case for stochastic gradient descent but I am getting the above error -- is there an import missing?

Base.:*(δη::Decay, x) = x/sqrt(δη.i+=1)

Thanks

Sven

It's at least as clear, faster, and more numerically stable.

very cool!
but to me, many of the images appear corrupted.

It might be helpful to explain the beauty taking place

What is the license for this code?

If you want people to be able to copy/paste this into projects, I would recommend putting the code in the public domain. Without a license, it cannot be used at all.

The implementation of the EM algorithm for GMMs is incorrect as it mixes univariate data x and multivariate component pdfs. Moreover, the calculations should be done in the log space and not in the space of probabilities for numerical reasons.

I believe the double sum in the E step should also be simplified to a single for loop.