Research & Insights
Sharing our contributions to the scientific community and the industry at large.
Featured Work
Next Engagement
Learning as Solving the Right PDE: Geometry, separability, curved manifolds and the structure hiding inside every learning problem
By: Alexandre Quemy
What if supervised learning were just solving the right partial differential equation? This talk traces the research journey behind Intrinsic Green's Learning — from a 2D Poisson classifier to Riemannian manifolds, brain-computer interfaces, and open problems at the frontier of geometry and machine learning.
Elements Scientific Conference
2026-10-23
All Talks & Presentations
Learning as Solving the Right PDE: Geometry, separability, curved manifolds and the structure hiding inside every learning problem
Presented by Alexandre Quemy
What if supervised learning were just solving the right partial differential equation? This talk traces the research journey behind Intrinsic Green's Learning — from a 2D Poisson classifier to Riemannian manifolds, brain-computer interfaces, and open problems at the frontier of geometry and machine learning.
Elements Scientific Conference
2026-10-23
Proofs: pivoting an agentic pre-seed startup
Presented by Alexandre Quemy
At Proofs, we operated a major shift from an autonomous multi-agents platform for end-to-end software development to human-in-the-loop multi-agents terminal-based tooling for software architecture.
We discuss a two years trajectory, from product market fit to technical and conceptual issues with agentic platforms.
AI SUMMIT BARCELONA
2026-09-15
LLM Function Calling: a Signature Model Approach
Presented by Alexandre Quemy
I present a client-side alternative to native function calling with many benefits in terms of cost, agentic capabilities and scalability.
Data Science Summit 2025 AI // ML Edition
2025-06-05
All Articles & Insights
Function Calling in LLMs: A Signature Model Approach
By: Alexandre Quemy
quemy.info - 2025-06-12
Introduces a novel approach to function calling in Large Language Models using a 'Signature Model' method. Proposes a client-side solution that overcomes current limitations in function calling.
Read Full Article →Stop using 'Correlation is not causation' and maybe stop using correlation
By: Alexandre Quemy
quemy.info - 2021-02-03
A critical examination of the concept of correlation, challenging the oversimplified "correlation is not causation" mantra and exploring the mathematical limitations of correlation coefficients.
Read Full Article →On controlling the propagation of numerical errors
By: Alexandre Quemy
quemy.info - 2021-01-05
Explores the challenges of numerical computations and floating-point arithmetic, focusing on how rounding errors propagate during calculations using the CESTAC method.
Read Full Article →Distributions, Laplace Transform and Initial Value Problem
By: Alexandre Quemy
quemy.info - 2021-01-09
Explores solving Initial Value Problems using distributions and the Laplace transform, presenting a unique algebraic perspective on the Laplace transform.
Read Full Article →Formal Publications
Filter by Year
2026
0 publicationsIntrinsic Green's Learning: Supervised Learning on Manifolds via Inverse PDE
2025
0 publicationsParkinson disease classification: a comparison of quantum and RBF kernels using support vector machine
The Riemannian Means Field Classifier for EEG-Based BCI Data
2024
0 publicationsEvaluation of the Electroencephalogram Conformer for the P300 Signal
Art makes quantum intuitive
Quantum Denoising in the Realm of Brain-Computer Interfaces: A Preliminary Study
Quantum machine-based decision support system for the detection of schizophrenia from EEG records
2023
0 publicationsA large reproducible benchmark on text classification for the legal domain based on the ECHR-OD repository
Judicial Independence and Impartiality: Tenure Changes at the European Court of Human Rights
MultiZenoTravel: a Tunable Benchmark for Multi-Objective Planning with Known Pareto Front
Case-Based and Quantum Classification for ERP-Based Brain–Computer Interfaces
First steps towards quantum machine learning applied to the classification of event-related potentials
Artificial intelligence and fair trial rights
ydata-profiling: Accelerating data-centric AI with high-quality data
Towards an architectural framework for intelligent virtual agents using probabilistic programming
pyRiemann-qiskit: a sandbox for quantum classification experiments with riemannian geometry
2021
0 publicationsCautiously Making Friends with AI: Machine Learning for human rights research and practice
Paradiseo: From a Modular Framework for Evolutionary Computation to the Automated Design of Metaheuristics
ECHR-OD: On Building an Integrated Open Repository of Legal Documents for Machine Learning Applications
2020
0 publicationsGBEx, towards Graph-Based Explanations
On Integrating and Classifying Legal Text Documents
Two-stage Optimization for Machine Learning Workflow
2019
0 publicationsBinary Classification In Unstructured Space With Hypergraph Case-Based Reasoning
Data Pipeline Selection and Optimization
Binary Classification With Hypergraph Case-Based Reasoning
2018
0 publicationsBinary Classification With Hypergraph Case-Based Reasoning
AI for the legal domain: an explainability challenge
Unsupervised Video Semantic Partitioning Using IBM Watson and Topic Modelling
2017
0 publicationsData Science Techniques for Law and Justice: Current State of Research and Open Problems
2015
0 publicationsSolving Large MultiZenoTravel Benchmarks with Divide-and-Evolve
True Pareto Fronts for Multi-objective AI Planning Instances
No publications found matching your search.
