Logic-Based Integrative Causal Discovery with Business Applications

Prof. Tsamardinos talk at LINCS (Laboratory of Information, Networking and Communication Sciences), Paris, France 24/02/2016.

Link to the video

Abstract: Computational Causal Discovery aims to induce causal models,causal networks, and causal relations from observational data withoutperforming or by performing only few interventions (perbutations,manipulations) of a system. While predictive analytics create models thatpredict customer behavior for example, causal analytics create models thatdictate how to affect customer behavior. A recent approach to causaldiscovery, which we call logic-based integrative causal discovery, will bepresented. This approach is more robust to statistical errors, makes morerealistic and less restrictive assumptions (e.g., admits latent confoundingfactors and selection bias in the data) and accepts and reasons withmultiple heterogeneous datasets that are obtained under different samplingcriteria, different experimental conditions (perbubations, interventions),and measuring different quantities (variables). The approach significantlyextends causal discovery based on Bayesian Networks, the simplest causalmodel available, and is much more suitable for real business or scientificdata analysis.

Integrative Causal Analysis paper accepted at KDD2016

The ETIO algorithm is a innovative causal analysis method that allows identifying causal relationships by integrating several datasets obtained under different experimental conditions. The algorithm will be presented in the prestigious ACM SIGKDD Conference on Knowledge Discovering and Data Mining (KDD 2016, 13 – 17 August).

KDD is an international event that brings together the top-players in the field of data mining, with a strong emphasis on applied data science.

Reference: G. Borboudakis and I. Tsamardinos. Towards Robust and Versatile Causal Discovery for Business Applications. 22nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2016).

Prof. Tsamardinos’s talk at the 9th Hellenic Conference on…

Prof. Tsamardinos  will present recent developments in logic-based causal discovery, based on the latest JMLR paper with Sofia Triantafillou:  “Constraint-based Causal Discovery from Multiple Interventions over Overlapping Variable Sets”   during the 9^th Hellenic   Conference on  Artificial Intelligence held in Grand Hotel Palace Thessaloniki, 18-20 May 2016.

Introducing Logic-based Integrative Causal Discovery

Prof. Tsamardinos will present the field of logic-based integrative causal discovery during two different talks to be held in Paris, France, in the upcoming week.

The talks will illustrate the characteristics of a new, logic-based approach to causal discovery, that has been partly pioneered by our group. This approach is more robust to statistical errors, makes more realistic and less restrictive assumptions, accepts and reasons with multiple heterogeneous datasets that are obtained under different sampling criteria, different experimental conditions, and measuring different quantities. The approach significantly extends causal discovery based on Bayesian Networks, the simplest causal model available, and is much more suitable for real business or scientific data analysis.

The two talks will be held in the premises of two prestigious research organizations:

• the Laboratoire de Recherche en Informatique (LRI), host Prof. Isabelle Guyon, on the 23-rd of February  (https://www.lri.fr)
• the Laboratory of Information, Networking and Communication Sciences (LINCS), host Prof. Dario Rossi, on the 24-th of February (LINCS talk)

MXM invited talk at ELIXIR meeting

ELIXIR is an inter-governmental organisation which builds on existing data resources and services within Europe. Its main objective is orchestrating the collection, quality control and archiving of large amounts of biological data produced by life science experiments.

Prof. Tsamardinos will give an invited talk at the forthcoming meeting “In silico infrastructures for Life Science’s Big Data – and the role of ELIXIR”, that will be held at the Hellenic Pasteur Institute,  Athens on February 26

The talk is entitled “Bioinformatics Advances for Automating Predictive Biomarker and Biosignatures from Data” and will focus on Biomarker and Biosignature discovery and on current tools for automatically discover them. Some of these tools have been realized by our group.

Find out more on the meeting website:

https://www.elixir-europe.org/events/infrastructures-life-sciences-big-data-and-role-elixir

New MXM website is up!

Our new web site is on-line! Check it out: www.mensxmachina.org