The 4th part of the hash map series is out:
I'm trying to keep explaining what I am doing while I hunt a bug from the resize! function

Paper writing pro tip, whenever you use a URL that is not permanent (a DOI), use the internet archive (archive.org) to ensure you are not citing a link that will be dead in two weeks from the publication.

The hash map video is uploading, I have to say video editing really is an underrated time-consuming task, thanks to content creators for putting up with it

Guess who learned the hard way to always check that your mic is on 🙃

I really enjoyed @jonhoo@twitter.com's stream on re-building the stdlib HashMap, I'll try doing a similar video with this weekend

One thing I wish it was was a completely self-contained, statically linked binary (in or ) instead of Python, I had to re-install it from scratch because of Python dependency hell

The papis command line is still a bit rough around the edges, but already extremely nice, the reference and document manager I've been searching for quite some time

From maths to code, manifold optimization with Optim.jl cc @pkofod@twitter.com

Dang, I hadn't had a white board talk for quite some time!

And now Antoine starts his talk on optimization on manifolds

TIL about Alloc.jl to turn functions into their in-place, allocation-free version

Filippo highlights the coolness of Julia for Physics applications, more specifically, NeuralQuantum.jl and quantum state inference with neural networks

And after PGMO and optimization, the Paris meetup @_Artelys@twitter.com

Is there a fix-point distribution wrt a given distributional Bellman operator?

Direct manipulation of probability distributions instead of expected values, a Bayesian touch in RL

Continuing the plenaries with distributional reinforcement learning

On the program, "it has evolved to include data scientists and the application of optimization to data science"

It takes this kind of conference to bring @Elvehen@twitter.com back to Europe 🙂

With the increase of electricity consumption, we have to ask ourselves how impacting each source is

A Mastodon instance for maths people. The kind of people who make $$\pi z^2 \times a$$ jokes. Use $$ and $$ for inline LaTeX, and $ and $ for display mode.