Machine Learning

Titanic Survival PredictionThis is the legendary Titanic ML competition — the best, first challenge for you to dive into ML competitions and familiarize yourself with how the Kaggle platform works.The competition is simple: use machine learning to create a model that predicts which passengers survived the Titanic shipwreck.The sinking of the Titanic is one of the most infamous … [Read more...] about Titanic Data Exploration

By: Brian Hie, Ellen D. Zhong, Bonnie Berger, Bryan Bryson Originally published in AAAS, Jan 15, 2021. Natural language predicts viral escape Viral mutations that evade neutralizing antibodies, an occurrence known as viral escape, can occur and may impede the development of vaccines. To predict which mutations may lead to viral escape, Hie et al. used a machine learning … [Read more...] about Learning the Language of Viral Evolution and Escape « Machine Learning Times

Imagine sending your 12 years old child to buy some milk at the supermarket near your house. You know that your neighbourhood is safe and you let him go alone. Unfortunately, something happens.A bad guy approaches your child trying to get him in his car: your child is about to be kidnapped. At the same moment, a patrol approaches and two police officers hear your child … [Read more...] about Predictive Policing: Effectiveness and Problems

Originally published in [email protected], Jan 19, 2021. When Facebook users scroll through their News Feed, they find all kinds of content — articles, friends’ comments, event invitations, and of course, photos. Most people are able to instantly see what’s in these images, whether it’s their new grandchild, a boat on a river, or a grainy picture of a band onstage. But many … [Read more...] about How Facebook is Using AI to Improve Photo Descriptions for People Who Are Blind or Visually Impaired « Machine Learning Times

Suis-je éthique ? “I’m sorry Dave, I’m afraid I can’t do that”Le 23 octobre 2020 le Parlement européen a émis des propositions et des recommandations pour un développement éthique de l’intelligence artificielle dans l’UE.Un des éléments clés de ces recommandations était d’éviter les biais et les discriminations des modèles de Machine Learning.Nous allons voir ici comment … [Read more...] about Ethique du Machine Learning : évaluer la “fairness” d’un modèle et le corriger

Originally published in UX Collective, Jan 19, 2021. For an industry that prides itself on moving fast, the tech community has been remarkably slow to adapt to the differences of designing with AI. Machine learning is an intrinsically fuzzy science, yet when it inevitably returns unpredictable results, we tend to react like it’s a puzzle to be solved; believing that with enough … [Read more...] about When Are We Going to Start Designing AI With Purpose? « Machine Learning Times

We built a machine model that predicted the Super Bowl would have record-low viewership. Here are three things it had to learn first.For any ambitious data scientist, the phrase “maybe this is a wild idea” is a clue to listen closely. Problem solving and creativity are among the most important traits of excellent data scientists, so you’re always listening for crazy ideas that … [Read more...] about Quality core data + non-parametric models are enabling previously impossible data projects

Why pretrained machine learning models are often unusable and irreproducible — and what we can do about it. Introduction A useful approach to designing software is through contracts. For every function in your codebase, you start by writing its contract: clearly specifying what inputs are expected and valid for that function (the precondition), and what the function will do … [Read more...] about Machine Learning Models are Missing Contracts « Machine Learning Times

The last hidden layer outputs 2 scalar latent features per node. These 2 scalar values represent a node. The nice part is, in this example, we can compute the latent representation based on the social graph alone even without training. In fact, we run the model for one iteration with W⁰ and W¹ initialized randomly with a normal distribution. Let’s not even perform any training … [Read more...] about Graph Convolutional Networks (GCN) & Pooling

Machine learning model performance often improves with dataset size for predictive modeling. This depends on the specific datasets and on the choice of model, although it often means that using more data can result in better performance and that discoveries made using smaller datasets to estimate model performance often scale to using larger datasets. The problem is the … [Read more...] about Sensitivity Analysis of Dataset Size vs. Model Performance