Since Artificial Intelligence (AI) algorithms have become popularized, especially by their employment in some of entertainment and social media’s most popular products or since they can rather successfully drive cars, they have come under public scrutiny like never before. People from all sorts of fields are, no doubted under the influence of celebrity scientists talking about AI, such as Elon Musk or Steven Hawking, wondering about the ethical implications of allowing algorithms to certain aspects of our lives and whether or not their career runs the risk of being replaced by the new technology. …

The analysis in the present report constitutes an overview of various efforts made in recent literature to apply Transformer based models to natural language processing tasks where meaning constitutes a difficult challenge. What we refer to with the latter will be later defined in the paper as “higher order” semantics, since the meaning of the sentences/language under consideration cannot be inferred solely from the mere morpho-syntactical composition. An analysis is also briefly given of the Transformer architecture in general (what sets it apart) and of BERT and XLNet, its two implementations most widely used in the papers surveyed. We propose…

Typical morning at work

We are more alike than we might think. Julian Baggini’s wonderful “How The World Thinks” notes that every nation displays every virtue and every vice. So too in philosophy, all cultures ask roughly the same questions and have the same preoccupations, amongst the most important of which are: theology, insight, logic, reason, pragmatism, tradition, time, naturalism, unity, reductionism, self, harmony, virtue, morality, liberty, transience, partiality, and belonging. …

Since 2017 a new architecture of Natural Language Processing (NLP) has succeeded in establishing itself as the state-of-the art technology for language related tasks: the transformer. The present paper provides a summary of the context in which this new model appear as well as some of the challenges that needed to be tackled. It then explains how the transformer helps solve them through the attention mechanism and mentions two of its more recent developments (2019 and 2020) and how they have surpassed previous benchmark scores.


Even though the transformer model as was as it was introduced in 2017 can be…

I don’t have enough room for my other stickers

The purpose of this article is to serve as a tiny handbook for anyone who already completed a Python tutorial, has a pretty good grasp on the basics and would like to take their skills to the next level. Becoming a good Python developer has a lot to do with getting yourself familiar with the language’s quirks and the pythonic way of thinking and doing things. I will go below through 6 points that will help a beginner-to-intermediate comfortably step into the domain of more advanced Python:

  • more complex list comprehensions
  • map, filter, and reduce with lambdas
  • locals and globals

Most of the tutorials and algorithms relating to solving Sudoku with a program will point you to a backtracking-implementing solution. Backtracking is an algorithm that recursively tries potential solutions and removes the ones that don’t work. Now, if the purpose is to learn and practice the algorithm, Sudoku will do perfectly fine, but, in many cases, backtracking has a high complexity and tends to reek of lack of properly understanding the problem due to its brute force approach. Let’s dive into how we can do it without!

Understanding the problem

In Sudoku, we need to fill the empty squares (which in our example…

Negoiţă D. D. Felix

Software Developer 💻🎧 #coding | Data Engineering Philologist and lover of #books 📚 | Tech enthusiast 📱 | @felixnego94

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