Links

Table of contents

Papers

• PhD - Statistical learning for selectivity estimation in relational databases (manuscript, slides)
• Selectivity correction with online machine learning - BDA 2020
• Selectivity Estimation with Attribute Value Dependencies using Linked Bayesian Networks - TLDKS 2020
• An Approach Based on Bayesian Networks for Query Selectivity Estimation - DASFAA, 2019
• Entropic Variable Projection for Explainability and Intepretability - 2018
• Master 2 year internship at HelloFresh (report, slides)
• Master 1 year internship at Privateaser (report, slides)
• Undergraduate internship at INSA Toulouse (report, slides)
• Detailed solutions to the first 30 Project Euler problems

Talks

• Manipulating ephemeral data with git - Défi IA 2021
• The challenges of online machine learning in production - Itaú Unibanco Meetup 2021
• Quelle est l’empreinte écologique du Big Data? - Toulouse Tech round table 2021
• A brief introduction to online machine learning - Hong Kong Machine Learning, 2020
• Online machine learning with decision trees - Toulouse AOC workgroup, 2020
• Our solution to the IDAO 2020 qualifiers - virtual seminar, 2020
• Global explanation of machine learning with sensitivity analysis - MASCOT-NUM, Paris, 2020
• The benefits of online learning - Quantmetry, Paris, 2019
• The benefits of online learning - Element AI, London, 2019
• The benefits of online learning - Airbus BizLab, Toulouse, 2019
• An approach based on Bayesian networks for query selectivity estimation - DASFAA, 2019
• Machine learning incrémental: des concepts à la pratique - Toulouse Data Science, 2019
• Online machine learning with creme - PyData, Amsterdam, 2019
• Docker for data science - HelloFresh, Berlin, 2017
• Challenge Big Data - Toulouse, 2017
• Forecasting bicycle-sharing usage - Toulouse Data Science, 2016

Datasets

• OpenBikes 2016 challenge

Blogroll

This is a list of blogs I regularly scroll through.

• Tim Salimans on Data Analysis
• Randal Olson
• Sam & Max – French and NSFW!
• Sebastian Raschka
• Clean Coder
• Pythonic Perambulations
• Erik Bernhardsson
• otoro
• Terra Incognita
• Real Python
• Airbnb Engineering
• No Free Hunch
• The Unofficial Google Data Science Blog
• will wolf
• Edwin Chen
• Use the index, Luke!
• Jack Preston
• Agustinus Kristiadi
• DataGenetics
• Katherine Bailey
• Netflix Research
• inFERENce
• Hyndsight – Rob Hyndman is a time series specialist.
• While My MCMC Gently Samples
• Ines Montani – By one of the founders of spaCy.
• Stephen Smerity
• Peter Norvig
• IT Best Kept Secret Is Optimization – By Jean-Francois Puget, aka CPMP.
• explained.ai
• Better Explained
• Genetic Argonaut
• pandas blog
• Towards Data Science
• Linear Digressions – data science podcasts.
• Not so standard deviations – more podcasts.
• Talking Machines – even more podcasts.
• Practical AI – here be podcasts.
• Probably Overthinking It
• Simply Statistics
• Practically Predictable
• koaning – By Vincent Warmerdam, who made calmcode.
• blogarithms
• Possibly Wrong
• FastML
• Parameter-free Learning and Optimization Algorithms
• Todd W. Schneider – This guy is really good at exploratory data analysis.
• Yann Thaddée – Not directly related to data science but interesting nonetheless.
• Colins Blog
• Fabien Sanglard – Nothing to do with data science, but such good taste!
• The Glowing Python – By the creator of MiniSom, which is worth checking out too.
• Matt Hancock
• Francis Bach – Someone with an h-index of 80+ who takes the time to blog is worth reading.
• Gwern Branwen – Cool in a weird way.
• Libres pensées d’un mathématicien ordinaire
• Count Bayesie
• Jim Savage
• Nick Higham – A lot of well explained algebra.
• Calmcode – Not a blog per se, but a nice collection of short to the point tutorials about various tools.
• Chris Said
• Evan Miller
• Eric Jang
• Andrey Akinshin
• Single Lunch
• Freakonometrics
• Martin Daniel
• Chris Kiehl
• ithaka.im – A guy I met who travelled for 6 years with his wife on a bike, very inspiring.
• Muthukrishnan – Has written some neat document processing stuff.
• Björn Ottosson
• Guilherme Duarte Marmerola
• Cal Paterson
• Claire Carroll

Hall of fame

The following is a hall of fame of papers, books, and blog posts that have a very high signal to noise ratio – at least in my book. I highly recommend reading some of them when you get time.

• The Elements of Statistical Learning - Jerome H. Friedman, Robert Tibshirani, and Trevor Hastie
• Machine Learning - Tom Mitchell – I think this wonderful textbook is under-appreciated.
• Artificial Intelligence: A Modern Approach - Russel & Norvig
• mlcourse.ai – Of all the introductions to machine learning, I think this is the one that strikes the best balance between theory and practice.
• Machine learning cheat sheets - Shervine Amidi
• Kalman and Bayesian Filters in Python - Roger Labbe – Kalman filters are notoriously hard to grok, this tutorial nicely builds up the steps to understanding them.
• CS231n Convolutional Neural Networks for Visual Recognition - Stanford
• Algorithmes d’optimisation non-linéaire sans contrainte (French) - Michel Bergmann
• Graphical Models in a Nutshell - Koller et al.
• Rules of Machine Learning: Best Practices for ML Engineering - Martin Zinkevich – You should read this once a year.
• A Few Useful Things to Know about Machine Learning - Pedro Domingos – This short paper summarizes basic truths in machine learning.
• Choose Boring Technology - Dan McKinley
• How to Write a Spelling Corrector - Peter Norvig – Magic in 36 lines of code.
• MCMC sampling for dummies - Thomas Wiecki
• Your Easy Guide to Latent Dirichlet Allocation
• An Intuitive Explanation of Convolutional Neural Networks - Ujjwal Karn
• An overview of gradient descent optimization algorithms - Sebastian Ruder
• How to explain gradient boosting - Terence Parr and Jeremy Howard – A very good introduction to vanilla gradient boosting with step by step examples.
• Why Does XGBoost Win “Every” Machine Learning Competition? - Didrik Nielsen – This Master’s thesis goes into some of the details of XGBoost without being too bloated.
• Good sleep, good learning, good life - Piotr Wozniak – Extremely long and nothing to do with data science, but a very thorough essay nonetheless on how to properly sleep.
• Make for data scientists - Paul Butler – I believe Makefiles are yet to be rediscovered for managing data science pipelines.
• Statistical tests, P values, confidence intervals, and power: a guide to misinterpretations – Just read it.
• The Cramér-Rao Lower Bound on Variance: Adam and Eve’s “Uncertainty Principle” - Michael Powers
• Kaggle contest on Observing Dark World - Cam Davidson-Pilon – If you’re not convinced about the power of Bayesian machine learning then read this and get your mind blown.
• A Concrete Introduction to Probability (using Python) - Peter Norvig – Extremely elegant Python coding.
• The Hungarian Maximum Likelihood Trick - Louis Abraham
• Machine Learning for Signal Processing - University of Illinois
• Don’t Call Yourself A Programmer, And Other Career Advice
• Tidy Data - Hadley Wickham – If you like playing with data then you need to be aware of this one.
• Gaussian Process, not quite for dummies - Yuge Shi – Gaussian processes are quite difficult to understand (at least, for me) but Yuge gives some great visual intuitions.
• Continuous Delivery for Machine Learning - Martin Fowler
• Memos - Sriram Krishnan
• Frequentism and Bayesianism: A Python-driven Primer - Jake VanderPlas
• A Few Useful Things To Know About Machine Learning - Pedro Domingos
• Multiworld Testing Decision Service: A System for Experimentation, Learning, And Decision-Making
• Machine Learning: The High-Interest Credit Card of Technical Debt - Google
• Variational Inference: A Review for Statisticians - David Blei and his flock
• The Performance of Decision Tree Evaluation Strategies - Andrew Tulloch
• Hidden Technical Debt in Machine Learning Systems - Google
• Distill: Why do we need Flask, Celery, and Redis? (with McDonalds in Between) - Lj Miranda – A good example of the difference between abstract ideas and implementation details.
• Darts, Dice, and Coins: Sampling from a Discrete Distribution - Keith Schwarz
• Simplifying Graph Convolutional Networks - Felix Wu et al. – A nice example of putting the horse before the cart.
• MIT 6.867 machine learning course notes - Tommi Jaakola – For people who enjoy concise mathematical notation.
• A Recipe for Training Neural Networks - Andrej Karpathy
• The Bitter Lesson - Richard Sutton
• The Best Medium-Hard Data Analyst SQL Interview Questions – There are some great interactive SQL tutorials out there, such as SQLBolt and Select Star SQL, but this one takes the cake due to its complexity.
• Rules of Programming - Rob Pike
• Transformers from scratch - Peter Bloem
• Understanding Matrix capsules with EM Routing - Jonathan Hui
• A Machine Learning Primer - Mihail Eric – A good read for beginners in machine learning algorithms.
• Fitting Bayesian structural time series with the bsts R package - Steven L. Scott
• Novelist Cormac McCarthy’s tips on how to write a great science paper - Savage and Yeh
• Emerging Architectures for Modern Data Infrastructure - Matt Bornstein, Martin Casado, and Jennifer Li – Gives a good overview of the data analysis tooling landscape as of late 2020.
• What your data team is using: the analytics stack - Technically – Another solid article to understand what an analytics stack looks like in 2021.
• Fred’s ImageMagick Scripts
• Unprojecting text with ellipses - Matt Zucker – See also this article on page dewarping by the same author.
• The Log: What every software engineer should know about real-time data’s unifying abstraction - Jay Kreps
• Language models, classification and dbacl - Laird A. Breyer – machine learning on text with a UNIX philosophy.
• Cameras and Lenses - Bartosz Ciechanowski – 100% worth a read.
• Ditherpunk - Surma
• Command-line Tools can be 235x Faster than your Hadoop Cluster - Adam Drake
• Teaching An Old Dog A New Trick - Chris Kamphuis
• Git scraping, the five minute lightning talk - Simon Willison – I wish I had thought about this first!
• Optimal Peanut Butter and Banana Sandwiches - Ethan Rosenthal
• The Data Science Hierarchy of Needs - Monica Rogati
• Are Pop Lyrics Getting More Repetitive? - Colin Morris
• Tuesday Changes Everything - Jesper Juul
• Gently down the stream - Mitch Seymour
• Les études statistiques sont-elles hors de contrôle? - David Louapre
• Visually stunning math concepts which are easy to explain - StackExchange
• Introduction to Locality-Sensitive Hashing - Tyler Neylon
• Transformers from scratch - Peter Bloem

Eye candy

• Tyler Hobbs – The god of generative arts.
• Some Jean Giraud stuff
• Mauro Martins
• A new way to knit by Petros Vrellis
• A fascinating article about Manolo Gamboa Naon
• Some Ukiyo-e
• Turtletoy
• Dwitter
• generated.space
• Pixel art by Marcus Blättermann
• Nick Barnes' football bible
• Simon Stålenhag
• Syd Mead (who worked on Blade Runner)
• Michael Fogleman’s blog
• World of Warcraft art by Dreamwalker
• Hors-sol de AKOREACRO
• Erica Anderson
• Jack Sharp
• Archillect – An AI that curates cool pictures, how awesome is that?
• Martin Kleppe
• Zoomquilt
• lossfunctions.tumblr.com – Yes, that’s a thing.
• Shirts of Peter Norvig
• United Airlines ads by Cream Electric Art
• Miniature Calendar by Tatsuya Tanaka – Broccolis that look like trees, staples that look like workout benches… I love it!