ARTIST
The project ARTIST was a 2022 and 2023 winner of the PREMATURATION call for projects, led by the Innovation Lab of IP Paris.
THE PROBLEM ADDRESSED
Today, unleashing AI performance is hurdled by:
Compliance (Ethics, Certification such as AI act)
Technics (unsupervised data, high dimension,...)
ARTIST tackles all that, with a SaaS solution and enables Data Team and End Users to build trustworthy AI (ethical, robust, explainable) to:
identify & correct biases and performance drift
provide top notch scoring performance
Powered by TreeRank algorithm +15y of R&D by Stéphan Clémençon (Telecom Paris patents, more than 30 rank A publications) teaming with a start up dedicated to explainable AI.
A top level scientific team (led by Stéphan Clémençon from Telecom Paris) and engineering team (AI-vidence start up and PhD from ENS).
TECHNOLOGY
The technology is based on the scoring method TreeRank, outperforming traditional techniques, (outcome of a long term research program in AI, by Prof. S. Clémençon, and partly with Prof. N. Vayatis (ENS Paris Saclay).
The algorithm is coded in Python, in a streamlit framework, hosted on Telecom Paris servers.
It may be additionally tweaked to enhance even the best performance achieved on any use case.
COMPETITIVE ADVANTAGES
- A solution to correct biases even in high dimension
- Optimal scoring model, enhancing the performance of any competitor
- A user-friendly interface with appealing interpretability/explicability features
- Simple to use, is fully explainable and requires no scientific expertise.