Alcides Fonseca

40.197958, -8.408312

CAMELOT: Autonomic platform for Machine Learning using anonymized data

CAMELOT is a CMU|Portugal joint project by Feedzai, CMU, U.Coimbra, IST and FCUL, funded by Lisboa2020, Compete2020 and FEDER.

I am the PI on the FCUL side, coordinating the efforts on verification and automatic optimization of machine learning pipelines.

The CAMELOT (autonomiC plAtform for MachinE Learning using anOnymized daTa) project aims at developing an innovative machine learning platform, which will tackle three key issues that hinder the efficiency and accuracy of modern AI applications:
• Ensuring real-time constraints during both the training and inference phases of machine learning models, while minimizing operational costs deriving from the use of cloud resources.
• Enabling learning over anonymized data, thus circumventing the privacy issues that currently prevent the reuse of information across models trained on datasets belonging to different entities (e.g., different financial institutions).
• Integrating information from different, independent and heterogenous data plataforms (e.g., key-value stores, relational and graph databases) in an automatic approach that maximizes the performance of machine learning applications.

CAMELOT was funded by some FCT, Lisboa2020, Portugal2020 and FEDER:

  • LISBOA-01-0247-FEDER-045915 and POCI-01-0247-FEDER-045915 – Main funding
  • CPCA/A1/402869/2021 – Camelot: Prediction of Machine Learning Pipeline Execution Times (2021-2022)
  • CPCA/A1/5613/2020 – Genetic Programming for Interpretability (2020-2021)
  • CPCA/A2/6009/2020CAMELOT Cloud (2020-2021)

More info

Team

Software

  • GeneticEngine An hybrid of Grammar-Guided and Strongly Typed Genetic Programming in Python.
  • Aeon A programming language with liquid types, focused on synthesis
  • LiquidJava A library+typechecker + VSCode plugin for Java that adds Liquid Types and TypeState.

Publications

Book Chapters
Journal
Conference Proceedings
Workshop Proceedings
Short Papers
Demos
Posters
Master Thesis

Funding

fct footer logo
fct footer logo
fct footer logo
fct footer logo
fct footer logo