Machine Learning Journal Club APS - University of Turin

About us

Read More


Read More

Learn with the MLJC

Read More


MLJC is Co-Host of Virtual BR41N.IO Hackathon – Spring School 2021


April 12-21, 2021


+3000 brain enthusiast



The Machine Learning Journal Club is honoured to be Co-Host of the Virtual Br41n Hackathon, organised by g.tec medical engineering in partnership with IEEE Brain.

The event will be held between the 17th and 18th of April 2021, as part of the BCI and Neurotechnology Spring School. MLJC will attend with five teams. We will have the opportunity to present our work and our association in front of an international audience.

All the results will be available open-source on our GitHub.

Brain Hackathon

BR41N.IO is a two days event designed to be a learning experience for developers, students, scientists and engineers who cram and build Brain Computer Interface (BCI) applications.

Spring School

A 10 days learning experince with Neurotechnology experts from all over the world. The members from MLJC are taking part to the Spring School organized by g.tec.

Our winning team

Our Teams

TOP-ECoG Hand Pose

Arianna Di Bernardo
Gabriele Penna
Simone Azeglio
Simone Poetto

Stroke Rehab Motor Imagery

Eleonora Misino
Lorenzo Migliori
Beatrice Villata
Pietro Sillano

P300-speller 1

Aurora Micheli
Flavio Sartori
Ilaria Gesmundo
Luca Bottero

P300-speller 2

Enrico Sansone
Pietro De Luca
Sara Giganti
Tanisha Khan

Syntetic SSVEP

Elios Ghinato
Giacomo D'Amicantonio
Pio Raffaele Fina
Marco Bottino
Vincenzo Triglione

Our teams at work

MLJC gets on the Neurohackathor 2021 podium

On 23rd and 24th May 2021 the Medical AI and BCI team from Machine Learning Journal Club participated in the NeuroHackathor 2021, hosted by the Neurotechnology Scientific Student Club from Torùn, Poland.

NeuroHackathor 2021 is an international hackaton for studying engineers, programmists, cognitive scientists and others interested in neurotechnology and human daily living.

The hackathon has been held online, and the Machine Learning Journal Club brought 11 components in 2 teams. Both teams have been awarded with second and third place!


MLJC team-1 choosed the “Museum of the future” topic. The aim was the personalization of the museum visitors’ experiences using technology. In 24h they developed an idea for an hybrid BCI-NLP-CV system to adapt the picture caption reading using multiple bio-signals as EEG, eye tracking and galvanic skin response.

Arianna Di Bernardo
Flavio Sartori
Gabriele Penna
Letizia Pizzini
Pio Raffaele Fina
Pietro Sillano

Here you can watch their pitch presentation:


MLJC team-2 choosed to work on the topic: “How to reduce the negative effects of pandemic isolation with neurotechnology?“. The main idea was to simulate the feeling of being in the nature, which is proven to be beneficial for depression, in a situation where a subject is forced in a closed space. They proposed a neurofeedback protocol and designed a minimalist & easy-to-use device, “The EGG”,  to perform the protocol.

Anjali Agarwal
Beatrice Villata
Enrico Sansone
Luca Pezzini
Marco Bottino
Micol Olocco
The EGG device

Here you can watch their presentation as well:

As every hackathon, it was an amazing intense experience but really fun, see you in the next one!

Politecnico di Torino on MLJC results at BCI Hackathon

MLJC team tops hackathon podium

MLJC is honoured to co-host BR4IN.IO hackathon on BCI

Presentations from AAAI-MLPS 2021 published

The MLJC has recently worked in the field of Scientific Machine Learning and its applications, thanks to various Julia libraries developed by Chris Rackauckas, who is contributing to the development of #SciML with his great work.
Therefore, we suggest listening to his amazing talk on Continuous-Time Echo State Nets, which you can find below.


MLJC presented in AAAI MLPS Symposium 2021

Simone Azeglio presented “Physics Informed Machine Learning Simulator for Wildfire propagation” at Mediterranean Machine Learning Summer School

Italian National television (TG1) reports about MLJC project “Physics-Informed Machine Learning Simulator for Wildfire Propagation”

Recruitment for Projects starting from Febr. 2021

We are opening the recruitment for our projects starting February 2021. The groups that are open for participation are:

  • Theoretical Machine Learning
  • Natural Language Processing and Social Sciences, 
  • Machine Learning for Health and Brain Computer Interface. 

These projects are intended to be both for beginner and experienced people. Beginners will have the opportunity to confront themselves with discussions with more experienced students. The final aim of these projects is to create innovative knowledge and materials, to participate in competitions, and to create a network of people with common interests.
You can find all the technical details of the projects here.

In order to participate to one or more of the groups, please fill the following form:

The deadline for the registration is the 15th February 2021.

In case of need of more information, feel free to contact us at:

We hope to hear from you soon!
The MLJC Team

About us

We are a group of students from the University of Turin and the Polytechnic University of Turin, mostly from a STEM background.

We are passionate about machine learning and artificial intelligence, thus we created a community where to share our geeky ideas and our expertise.

We are actively engaged in research projects with a broad, interdisciplinary focus and participating in hackathons / competitions related to AI.

If you’re interested in taking part in any way in our research projects or learning activities, feel free to drop an email at:

What we do

The aim of our association is to promote hands-on knowledge of machine learning and artificial intelligence across the academic and civic landscape of Turin.

We believe interdisciplinarity and cross-contamination can be the key to new discoveries and fruitful, engaging ideas.

We try to innovate learning and research paradigms with a focus on self-organization and peer-learning / peer-reviewing, ultimately producing industry-grade science.

In order to promote machine learning literacy we jump-started a wide set of projects, on various premises and online:

  • We took over the University of Turin after-hours, delivering free-for-all courses on Python for machine learning
  • We engaged our more experienced members in open and horizontal research projects on natural language processing, theoretical machine learning and medical machine learning
  • We encouraged and financed the participation of our members to hackathons and competitions
  • We published divulgatory articles on machine learning techniques and hot topics

Our Story

We started our first activities as a little group of students from the M.Sc. in Physics of Complex Systems at the University of Turin.
We were curious about emergent possibilities stemming from Data Mining, Machine Learning and Artificial Intelligence: that’s why we started experimenting with real-world projects and participating in some Kaggle competitions we were passionate about (for example, we took part in the Deep Fake Detection Challenge).

Going forward, we decided to open our group to students from different backgrounds: in order to build some common knowledge, we started organizing extra-curricular lessons and activities, covering Python applied to machine learning (you can find the course material here).

As we meet and engage with new people from all walks of science, our community grows large and diverse: we are proud to count members from computer science, engineering, mathematics, biology, economics and neuropsychology to name a few.

We meet regularly (online or in person), to work and organize projects related to Machine Learning and its applications. You can find more information about our projects here.

MLJC presented “Physics Informed Machine Learning Simulator for Wildfire Propagation” at AAAI Symposium 2021

MLJC presented the following research at AAAI-MLPS Symposium 2021 on February 2021. Please find the proceedings below or in the official website:

The aim of this work is to evaluate the feasibility of re-implementing some key parts of the widely used Weather Research and Forecasting WRF-SFIRE simulator by replacingits core differential equations numerical solvers with state-of-the-art physics-informed machine learning techniques tosolve ODEs and PDEs, in order to transform it into a real-time simulator for wildfire spread prediction. The main programming language used is Julia, a compiled language which offers better perfomance than interpreted ones, providing Justin Time (JIT) compilation with different optimization levels. Moreover, Julia is particularlywell suited for numerical computation and for the solution ofcomplex physical models, both considering the syntax and thepresence of specific libraries such as DifferentialEquations.jl and ModelingToolkit.jl.

Our toolbox