Random stuff
I’ve been listening to some podcasts lately and I thought it would be dope to have them summarized chatgpt-style after each episode. Unfortunately I wasn’t super convinced with the free solutions I’ve found online, either YouTube wasn’t supported out of the box or the summary was meh. I don’t think it’s that hard to build so let’s try
A friend found this nice website deep-ml/ which attempts to re-create the “Leetcode experience” but for Machine Learning / Data Science. I’ll try to solve some of these / write some refreshers on the topics. Problems Exercice 1 Exercice 2 Exercice 3 Matrix times a Vector Write a Python function that takes the dot product of a matrix and a vector. return -1 if the matrix could not be dotted with…
Disclaimer (this is financial advice)
Probabilities can be confusing, and I’m still wrapping my head around some (simple) concepts. So… Random Variables are NOT Random The term “random variable” is actually a misnomer, it is formally defined as a deterministic function that takes as input any element from the sample space and returns a number from the set of possible results of the experiment (the sample space \(\Omega\) is…
Bridgewater (4% returns) vs index funds (18%). I'm not saying you can't beat the market. But Bridgewater is a hedge fund with 2,000 employees. Some of the smartest, highest paid people on earth who spend 20 hours a day + billions on tools/research. And they, and many just… pic.twitter.com/YKLzZaikbb — Sam Parr (@thesamparr) October 7, 2024 I’ve seen this tweet recently comparing returns of…
Why ? Because it’s funny How ? The whole thing is actually quite simple with Google Face Landmarker , which is basically 3 models packaged together: Face detection model : detects the presence of faces with a few key facial landmarks Face mesh model: adds a complete mapping of the face. The model outputs an estimate of 478 3-dimensional face landmarks. Blendshape prediction model: receives output…
Firms love to force candidates to go through bizarre humiliation rituals during interviews. One of these was to implement a topsort algo for their autograd engine. If we consider a computation graph where each node represents a variable and each edge represents a dependency, the goal of this algorithm is to sort the nodes in “ topological order ”, i.e. such that if there is an edge from node \(u\)…
What are we doing here ? Yeah, so recently I’ve been reading “ The Man Who Solved The Market ” by Gregory Zuckerman . Fun book about Jim Simons and his famously cracked Renaissance Technologies hedge fund. Unfortunately, it didn’t make me better at trading lol Use my ref link 👽 BUT it motivated me to learn more about some of the maths they used. Notably, Markov chains are mentioned a few times,…
Page en léger, où le but est de résoudre les problèmes du famoso PDF LLG . J’essaie quand c’est possible de donner une solution différente de celle donnée dans le corrigé (semi) officiel et de faire le lien avec d’autres notions ou problèmes. Liste des problèmes Exercice 174 Exercice 266 Exercice 287 Exercice 318 Exercice 389-390-392-393 Exercice 391 Interlude Facile d’accès Exercice 174 (Une…