E-values, a better alternative for p-values — In scientific research, the manipulation of data and peeking at results have been problems for as long as the field has existed. Researchers often aim for a significant p-value to get published, which can lead to the temptation of stopping data collection early or manipulating the data. This practice, known…

Delve into the motivations and consequences of p-hacking — A recent New Yorker headline reads, “They Studied Dishonesty. Was Their Work a Lie?”. What’s the story behind it? Behavioral economist Dan Ariely and behavioral scientist Francesca Gino, both acclaimed in their fields, are under scrutiny for alleged research misconduct. To put it bluntly, they’re accused of fabricating data to…

Discover 5 inspiring topics empowered by artificial intelligence — There are many examples of using AI in the wrong way, as highlighted in the thought-provoking book Weapons of Math Destruction. Also, the risks of AI should not be underestimated. AI ethics and governance have become pressing concerns in our rapidly evolving technological landscape, and numerous companies now maintain dedicated…

Solving optimization problems and enhancing results with ACO in Python — Welcome back! In my previous post, I introduced the fundamentals of Ant Colony Optimization (ACO). In this installment, we’ll delve into implementing the ACO algorithm from scratch to tackle two distinct problem types. The problems we’ll be addressing are the Traveling Salesman Problem (TSP) and the Quadratic Assignment Problem (QAP)…

An introduction to a lesser-known heuristic based on the behavior of ants — In the world of optimization algorithms, there are a plethora of methods inspired by the wonders of the natural world. From genetic algorithms based on evolution to the cooling strategies of simulated annealing, these algorithms have demonstrated their efficacy in solving complex problems. However, nestled in this diverse landscape of…

An introduction to one of the central questions in computer science — How come the shortest path problem is easy to solve, but the traveling salesman problem is not? What are the mathematical ideas on this? How to determine if a problem will take an unmanageable number of steps if its size increases? In this post you will learn the basics on…

Advanced coding tips for data scientists — Unlike traditional software engineering projects, ML codebases tend to lag behind in code quality due to their complex and evolving nature, leading to increased technical debt and difficulties in collaboration. Prioritizing maintainability is important to create robust ML solutions that can adapt, scale, and deliver value over time. In recent…

An introduction to graph theory and its applications — In this post, we delve into the world of mathematical optimization within graphs, exploring key concepts, algorithms, and practical applications. Graph problems can be found in many places. Obvious ones are in logistics or social network analysis, like finding the optimal route for a delivery company or the lowest amount…

Increasing the chances of getting a model to production — When evaluating and comparing machine learning solutions, your first go-to evaluation metric will probably be predictive power. It’s easy to compare different models with one single metric, and this is perfectly fine in Kaggle competitions. In real life, the situation is different. Imagine two models: one model that uses 100…