Sign in

Machine Learning @ University of Cambridge | École Polytechnique.

Are two objects that share the same properties the same object?

Photo by Constantinos Kollias on Unsplash


The Principle Identity of Indiscernibles (PII) states, in its least formal form, that two things that share all of their properties are in fact one and the same thing. It is the converse of the undisputed Principle of Indiscernibility of Identicals which states, intuitively, that two identical things cannot differ in any way. When combined, the two are sometimes referred to as Leibniz’s Law.

The PII has served as an important metaphysical axiom since the 1800s and is quite uncontroversial when applied to the objects with which we interact on a daily basis (trees, rocks, etc.). However, as we would…

A rhetorical analysis of Ray Bradbury’s novel, Fahrenheit 451.

Photo by Jay Heike on Unsplash


“It was a pleasure to burn” [1]. Ray Bradbury’s dystopian novel, Fahrenheit 451, offers readers a vivid account of how life could become if a technologically advancing society chose to value instant-gratification and ignorance over passion and knowledge.

The story centers around the life of the protagonist, Guy Montag, and his intellectual rebellion against a mindless and deceitful society focused primarily on efficiency and the creation of like-minded, brainwashed individuals. …

A thorough introduction to some of the newest algorithms in policy evaluation, including GTD and GTD2.

Photo by Rafif Prawira on Unsplash

Table of Contents:

  • Introduction
  • Linear Function Approximation
  • Deriving the GTD2 algorithm
  • Conclusion
  • Acknowledgements and Resources


Reinforcement learning is one of the hottest fields to be in right now, with concrete applications growing at an incredibly rapid pace, from beating video games to robotics. At its essence, reinforcement learning (RL) deals with decision making —i.e. it attempts to answer the question of how an agent should act in a given environment.

Reinforcement learning deals with decision making

Loosely speaking, all of RL comes down to either finding or evaluating a policy, which is just a way of behaving. …

An introduction to the Cramer-Rao lower bound for the variance/MSE of unbiased estimators

Photo by Macau Photo Agency on Unsplash

Introduction — What is an estimator?

Suppose you’re flipping a coin and you want to empirically estimate the probability of getting heads. To do this, you decide to flip your coin n times and record a one or a zero depending on whether you get heads or tails.

The most straightforward way to estimate the probability would be to compute the proportion of heads that you obtained in your n throws, which we’ll call p̂. The bigger n gets, the more certain you are that accurately estimates p, the true probability of getting heads. In statistics, we say that is an estimator of p

An introduction to some of the most fundamental concepts in logic programming (and constraint logic programming) with Prolog.

Image Source: Janeb13


If you’ve coded before, chances are you’re familiar with an imperative language, such as Python, Java, or C++. In this paradigm, a program is a list of instructions that modify its state when executed. Although this is the most common way of programming, it isn’t the focus of this article.

Instead, we’re going to introduce a different programming paradigm, logic programming, wherein a program is a database of relations. We will lay out the main concepts as we try to solve some simple questions in one of the most popular logic languages, prolog.

Working with Relations — is Socrates mortal?

One of the most fundamental concepts in…

Using decision trees to approximate the year that a book was written

Photo by Kiwihug on Unsplash


What if we used machine learning to give an estimate for when the book was written?

Picture yourself in your grandparents’ old attic, perusing the old, dusty bookshelf when an unfamiliar tome grabs your eye. It appears to be a novel. You don’t know when it was written, an answer that even seems to elude Google itself. You snap a picture of one of the pages and, using a text extractor, you obtain a digital representation of the text. What if we used machine learning to give an estimate for when the book was written?

In this post, we’re going…

A practical introduction to incremental learning in Python using scikit-multiflow



Data is all around us. Whether it’s profile pictures, tweets, sensor applications, credit card transactions, emails, or news feeds, data is here…and it’s being generated at incredibly fast speeds. With these seemingly infinite streams of data, one of the key challenges is to create lightweight models that are always ready to predict and adaptive to changes in the data distribution. The limitations of traditional machine learning methods in this setting has led to the development of online learning (also called incremental learning) methods.

In this post, we will gently introduce incremental learning through a practical implementation of a simple online…

Louis de Benoist

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store