This month: Facebook AI vs brain signals 👍, dendritis in artificial neural networks 🚨, understanding illusions in retina dynamics 👁️

Image by Federico Beccari on Unsplash

Deep recurrent Encoder: A scalable end-to-end network to model brain signals

Omar Chehab, Aleandre Defossez, Jean-Christophe Loiseau, Alexandre Gramfort, Jean-Remi King, Paper, Code

This paper comes directly from a collaboration between Facebook AI Research, the University Paris-Saclay and Ecole Normal Superieure. The main key of the paper is to devise a new method to help the neuroscience community to analyse brain responses to sensory inputs. …

Every day hundreds of pre-reviewed papers are published on Stay updated with me, with this mini guide about the most interesting papers for each month. The most peculiar papers for February 2021

The brain. Image by Alina Grubnyak on Unsplash

Neuroscience is the root for nowadays artificial intelligence 🧠🤖. Reading and being aware of the evolution and new insights in neuroscience not only will allow you to be a better “Artificial Intelligente” guy 😎, but also a finer neural network architectures creator 👩‍💻!

February 2021 has 3 interesting papers which are worth a read, here I summarize and convert them to layman terms

The geometry of information coding in correlated neural populations

Rava Azeredo da Silveira…


This is my third tutorial on Cython. This time we are going to implement the Detrended Fluctuation Analysis (DFA), a widely used technique for time series analysis, ranging for music to finance

Can we unveil the hidden nature of long time events? Image by Isaac Smith on Unsplash

Welcome back to my Cython tutorials! Today we will be dealing with a widely use technique, called Detrended Fluctuation Analysis (DFA). DFA has been intensively applied in music¹ ² and finance³, being able to capture correlations trends in time series and non-stationary signals — namely signals changing in time. The idea of this analysis methodology is simple and straight and well founded mathematically since the ‘50's. As…

Hands-on Tutorials

This is my second tutorial on Cython. This time I am going to introduce you the Non Negative Matrix Factorisation (NMF) and show its applications to music and NLP

Underlying architectural motifs, (de)composition. Image by Jung Ho Park

Welcome back to the Cython world :) This time I will show you how to implement a basic version of non-negative matrix factorisation (NMF). NMF has wide applications in data science¹ ² ³, music and astronomy, and its implementation gives us the opportunity to take a look at memory-views memory management in Cython.

Thus if you are willing to know about how NMF works and how to write it up in Python, C and Cython, you are in the right place. All the codes needed for this tutorial can be found here:

These codes are extremely useful to understand…

Disunited States of America, hate, and bots from Instagram comments

How are Americans living the electoral campaign? I scraped more than 400'000 comments from Trump’s and Biden’s Instagram feed and analysed them. A fragmented America emerges, with people looking for new hope and leadership and people (or bots?) covering with mud political opponents

Image by Kat Combs

On the 3rd November the 46th president of the United State of America will be elected. The main fight is between the current republican president Donald J. Trump and the ex Obama’s vice president Joe R. Biden. Analysts predict Biden to win, while Trump’s odds are about 1/10, which are not so high but still statistically significant…

Do you want to know more about Cython? Follow me in this series and I will show you practical examples of C-Cython-Python implementations

Spectrogram of “Compression of Time”, Final Fantasy 8

I love so much Cython, as it takes the best of the two main programming worlds: C and Python. Both languages can be combined together in a straightforward way, in order to give you more computational efficient APIs or scripts. Furthermore, coding in Cython and C helps you to understand what is underneath common python packages such as sklearn, bringing data scientists to a further step, which is quite far from the simple import torch and predefined algorithms usage.

In this series I am going to show you how to implement in C and Cyhton algorithms to analyse music, following…

Ciao a tutti! Questa story sara’ totalmente in italiano e avra’ a che fare con un famosissimo duo/team di Youtube: The Show.

Recentemente i The Show hanno dovuto rivedere la loro programmazione, a causa del tristemente noto COVID-19. In uno dei video piu recenti han proposto di rispondere ai loro haters. Sinceramente non so come i The Show possano navigare tra i loro commenti e se Youtube stesso dia la possibilita di avere una navigazione facilitata, in ogni caso son rimasto sorpreso nel vedere che Valerio si sarebbe accollato il lavoraccio di scrollare tutti i commenti ricevuti. Questo video mi…

In this second post I will extend the formalism introduced previously for a general exponential family distribution for the outcome Y and I will derive the basic equations to introduce and solve the maximum likelihood estimation. This post is quite mathematical heavy, but keep on with this as in the third part we will get our hands dirty in implementing a linear regression in different languages.

One of the main assumption to derive a linear model is that the outcome Y belongs to an exponential family distribution, whose general form is:

In this series of posts I am going to show applications and practical example on how to use Scala and its functionalities. In particular, all the examples are taken from this book: Scala Machine Learning Projects.
I found Scala ML Projects a very interesting reading and I think it’s good to keep up an updated repository, where users can easily retrieve input files and being up to date with the current Scala updates.

I am currently running these notebooks with Scala 2.12. If you want to run them online you can use almond-scala. Almond is a fantastic project, which keeps…

I will try to review and extend more and more this post, in order to show the power and importance of Generalised Linear Model and giving clear practical examples

Generalised Linear Model or GLM are a vast class of models, which try to fit a distribution of points (observations), independently from the distribution function of the observations under study. Such a statement sounds quite strong, but it has remarkable applications from biology to finance, creating a solid foundations for nowadays machine learning world. We often use models derived from GLM in our data science projects (e.g. Linear Regression, Logistic Regression…

Stefano Bosisio

Data Scientist with a PhD in Computational Chemistry. Passionate about AI and coding.

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