At, Research in AI and ML isn't just a goal — it's our essence

Rooted in academic rigor, our team is devoted to expanding the frontiers of knowledge and spearheading novel research initiatives in collaboration with some of the world's leading academic institutions. We are engaged in a dual mission: advancing near-term, product-focused innovations and pioneering long-term developments destined to redefine the technology landscape. At, we're not just part of the future of AI and ML—we're actively creating it.

conceptual image of brain connections
diagram explanation of AGI research. Explore, think and create.

Leading the Charge in AGI: Explore, Think, Create

At, our vision is driven by the groundbreaking pursuit of Artificial General Intelligence (AGI) — a journey that holds the promise of revolutionizing the realm of intelligent systems. At the heart of our endeavors, we are crafting a new class of algorithms, inspired by the human brain's complexity, to push the boundaries of what AI can achieve. Our work initially concentrates on elevating abstract reasoning, merging advanced mathematics, logic, and cutting-edge deep language models into integrated, high-performance framework.

Our mission is to nurture a form of machine intelligence that stands out for its adaptability and efficiency. Drawing on neuroscience's profound insights and the leaps in computational theory, we are on a mission to carve a pathway toward an era where AGI can address diverse challenges with unmatched finesse. By optimizing computational resources and harmonizing various learning paradigms, our goal is to see AGI not just as a tool, but as a transformative force that amplifies human achievement and fundamentally alters our societal landscape.

Our Current Research Initiatives

Investigating cognitive architectures

Exploring various cognitive frameworks to enhance AI’s decision-making capabilities and ensure it can operate transparently and effectively.

Learning through algebraic bases

Creating new learning architectures by leveraging advanced algebraic structures with the goal of representing information in a more efficient way.

Topology of dynamic graphs and multi-variate time series

Experimenting with a graph-based approach to provide an interpretable and structured format for intuitive content navigation and addressing the challenges of data retrieval.

Open Source Contributions

At we believe in the power of open collaboration to drive innovation in AI and ML.
Our commitment to advancing the field is reflected in our contributions to open source, where we
share tools and libraries designed to solve complex problems and inspire new ideas.


A high-performance topological machine learning toolbox in Python, designed to bring topological analysis to the forefront of machine learning research and application.


Deep learning meets topology, offering tools to incorporate topological data analysis into deep learning frameworks, enhancing model interpretability and performance.


An all-encompassing time series analysis suite, providing robust tools for forecasting, anomaly detection, and feature extraction in temporal data.


A high-performance implementation of Vietoris-Rips persistence, enabling efficient computation of persistent homology, a cornerstone of topological data analysis.


Essential distribution tools for scaling deep learning computations, facilitating the development and deployment of AI models across diverse computational environments.

Our Research Publications is committed to advancing the field of AI and ML. Our research team regularly publishes
findings, methodologies, and insights that contribute to the broader scientific community.

April 2024

Detection and Grading of Radiographic Hand Osteoarthritis Using an Automated Machine Learning Platform


Leo Caratsch, Christian Lechtenboehmer, Matteo Caorsi, Karine Oung, Fabio Zanchi, Yasser Aleman, Ulrich A. Walker, Patrick Omoumi, Thomas Hügle

Short Abstract / Summary

Automated machine learning (autoML) platforms allow health care professionals to play an active role in the development of machine learning (ML) algorithms according to scientific or clinical needs. The aim of this study was to develop and evaluate such a model for automated detection and grading of distal hand osteoarthritis (OA).

March 2023

Early Warning Signals of Social Instabilities in Twitter Data


Vahid Shamsaddini, Henry Kirveslahti, Raphael Reinauer, Wallyson Lemes de Oliveira, Matteo Caorsi, Etienne Voutaz

Short Abstract / Summary

The goal of this project is to create and study novel techniques to identify early warning signals for socially disruptive events, like riots, wars, or revolutions using only publicly available data on social media. Such techniques need to be robust enough to work on real-time data: to achieve this goal we propose a topological approach together with more standard BERT models. Indeed, topology-based algorithms, being provably stable against deformations and noise, seem to work well in low-data regimes. The general idea is to build a binary classifier that predicts if a given tweet is related to a disruptive event or not. The results indicate that the persistent-gradient approach is stable and even more performant than deep-learning-based anomaly detection algorithms. We also benchmark the generalisability of the methodology against out-of-samples tasks, with very promising results.

August 2021

ICLR 2021 Challenge for Computational Geometry & Topology: Design and Results


Nina Miolane, Matteo Caorsi, Umberto Lupo, Marius Guerard, Nicolas Guigui, Johan Mathe, Yann Cabanes, Wojciech Reise, Thomas Davies, António Leitão, Somesh Mohapatra, Saiteja Utpala, Shailja Shailja, Gabriele Corso, Guoxi Liu, Federico Iuricich, Andrei Manolache, Mihaela Nistor, Matei Bejan, Armand Mihai Nicolicioiu, Bogdan-Alexandru Luchian, Mihai-Sorin Stupariu, Florent Michel, Khanh Dao Duc, Bilal Abdulrahman, Maxim Beketov, Elodie Maignant, Zhiyuan Liu, Marek Černý, Martin Bauw, Santiago Velasco-Forero, Jesus Angulo, Yanan Long

Short Abstract / Summary

This paper presents the computational challenge on differential geometry and topology that happened within the ICLR 2021 workshop "Geometric and Topological Representation Learning". The competition asked participants to provide creative contributions to the fields of computational geometry and topology through the open-source repositories Geomstats and Giotto-TDA. The challenge attracted 16 teams in its two month duration. This paper describes the design of the challenge and summarizes its main findings.

March 2021

A Topological Data Analysis Toolkit for Machine Learning and Data Exploration


Guillaume Tauzin, Umberto Lupo, Lewis Tunstall, Julian Burella Pérez, Matteo Caorsi, Wojciech Reise, Anibal Medina-Mardones, Alberto Dassatti, Kathryn Hess

Short Abstract / Summary

We introduce giotto-tda, a Python library that integrates high-performance topological data analysis with machine learning via a scikit-learn-compatible API and state-of-the-art C++ implementations. The library's ability to handle various types of data is rooted in a wide range of preprocessing techniques, and its strong focus on data exploration and interpretability is aided by an intuitive plotting API. Source code, binaries, examples, and documentation can be found at

March 2024

Graph Language Model (GLM): A new graph-based approach to detect social instabilities


Wallyson Lemes de Oliveira, Vahid Shamsaddini, Ali Ghofrani, Rahul Singh Inda, Jithendra Sai Veeramaneni, Étienne Voutaz

Short Abstract / Summary

This scientific report presents a novel methodology for the early prediction of important political events using News datasets. The methodology leverages natural language processing, graph theory, clique analysis, and semantic relationships to uncover hidden predictive signals within the data. Initially, we designed a preliminary version of the method and tested it on a few events. This analysis revealed limitations in the initial research phase. We then enhanced the model in two key ways: first, we added a filtration step to only consider politically relevant news before further processing; second, we adjusted the input features to make the alert system more sensitive to significant spikes in the data. After finalizing the improved methodology, we tested it on eleven events including US protests, the Ukraine war, and French protests. Results demonstrate the superiority of our approach compared to baseline methods. Through targeted refinements, our model can now provide earlier and more accurate predictions of major political events based on subtle patterns in news data.

December 2021

Persformer: A Transformer Architecture for Topological Machine Learning


Raphael Reinauer, Matteo Caorsi, Nicolas Berkouk

Short Abstract / Summary

One of the main challenges of Topological Data Analysis (TDA) is to extract features from persistent diagrams directly usable by machine learning algorithms. Indeed, persistence diagrams are intrinsically (multi-)sets of points in R2 and cannot be seen in a straightforward manner as vectors. In this article, we introduce Persformer , the first Transformer neural network architecture that accepts persistence diagrams as input. The Persformer architecture significantly outperforms previous topological neural network architectures on classical synthetic and graph benchmark datasets. Moreover, it satisfies a universal approximation theorem. This allows us to introduce the first interpretability method for topological machine learning, which we explore in two examples.

August 2021

A Python Library for High-Performance Computation of Persistent Homology of Vietoris-Rips Filtrations


Julián Burella Pérez, Sydney Hauke, Umberto Lupo, Matteo Caorsi, Alberto Dassatti

Short Abstract / Summary

We introduce giotto-ph, a high-performance, open-source software package for the computation of Vietoris-Rips barcodes. giotto-ph is based on Morozov and Nigmetov's lockfree (multicore) implementation of Ulrich Bauer's Ripser package. It also contains a re-working of the GUDHI library's implementation of Boissonnat and Pritam's Edge Collapser, which can be used as a pre-processing step to dramatically reduce overall run-times in certain scenarios. Our contribution is twofold: on the one hand, we integrate existing state-of-the-art ideas coherently in a single library and provide Python bindings to the C++ code. On the other hand, we increase parallelization opportunities and improve overall…

Let's innovate together

Contact us to discuss how we can work together to explore the potential of AI for achieving your goals.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.