no code implementations • 22 Feb 2024 • Konstantina Biza, Antonios Ntroumpogiannis, Sofia Triantafillou, Ioannis Tsamardinos
We introduce the concept of Automated Causal Discovery (AutoCD), defined as any system that aims to fully automate the application of causal discovery and causal reasoning methods.
1 code implementation • 11 Dec 2023 • Giorgos Borboudakis, Paulos Charonyktakis, Konstantinos Paraschakis, Ioannis Tsamardinos
AutoML platforms have numerous options for the algorithms to try for each step of the analysis, i. e., different possible algorithms for imputation, transformations, feature selection, and modelling.
1 code implementation • 13 Sep 2022 • Antonios Ntroumpogiannis, Michail Giannoulis, Nikolaos Myrtakis, Vassilis Christophides, Eric Simon, Ioannis Tsamardinos
The behavior of the detectors is correlated with the characteristics of different datasets (i. e., meta-features), thereby providing a meta-level analysis of their performance.
no code implementations • 18 Oct 2021 • Nikolaos Myrtakis, Ioannis Tsamardinos, Vassilis Christophides
PROTEUS is designed to return an accurate estimate of out-of-sample predictive performance to serve as a metric of the quality of the approximation.
no code implementations • 9 Dec 2020 • Anastasios Tsourtis, Yannis Pantazis, Ioannis Tsamardinos
Inferring the driving equations of a dynamical system from population or time-course data is important in several scientific fields such as biochemistry, epidemiology, financial mathematics and many others.
no code implementations • 1 Apr 2020 • Michail Tsagris, Zacharias Papadovasilakis, Kleanthi Lakiotaki, Ioannis Tsamardinos
Feature selection for predictive analytics is the problem of identifying a minimal-size subset of features that is maximally predictive of an outcome of interest.
no code implementations • 23 Aug 2017 • Ioannis Tsamardinos, Giorgos Borboudakis, Pavlos Katsogridakis, Polyvios Pratikakis, Vassilis Christophides
We present the Parallel, Forward-Backward with Pruning (PFBP) algorithm for feature selection (FS) in Big Data settings (high dimensionality and/or sample size).
no code implementations • 23 Aug 2017 • Ioannis Tsamardinos, Elissavet Greasidou, Michalis Tsagris, Giorgos Borboudakis
BBC-CV's main idea is to bootstrap the whole process of selecting the best-performing configuration on the out-of-sample predictions of each configuration, without additional training of models.
no code implementations • 30 May 2017 • Giorgos Borboudakis, Ioannis Tsamardinos
In experiments we show that the proposed heuristic increases computational efficiency by about two orders of magnitude in high-dimensional problems, while selecting fewer variables and retaining predictive performance.
no code implementations • 10 Nov 2016 • Vincenzo Lagani, Giorgos Athineou, Alessio Farcomeni, Michail Tsagris, Ioannis Tsamardinos
The statistically equivalent signature (SES) algorithm is a method for feature selection inspired by the principles of constrained-based learning of Bayesian Networks.
no code implementations • 9 Aug 2014 • Giorgos Borboudakis, Ioannis Tsamardinos
A significant theoretical advantage of search-and-score methods for learning Bayesian Networks is that they can accept informative prior beliefs for each possible network, thus complementing the data.
no code implementations • 15 Apr 2014 • Christina Papagiannopoulou, Grigorios Tsoumakas, Ioannis Tsamardinos
Marginal probabilities are entered as soft evidence in the network and adjusted through probabilistic inference.
no code implementations • 10 Mar 2014 • Sofia Triantafillou, Ioannis Tsamardinos
In this work, we present algorithm COmbINE, which accepts a collection of data sets over overlapping variable sets under different experimental conditions; COmbINE then outputs a summary of all causal models indicating the invariant and variant structural characteristics of all models that simultaneously fit all of the input data sets.
no code implementations • 28 Sep 2012 • Giorgos Borboudakis, Ioannis Tsamardinos
A significant theoretical advantage of search-and-score methods for learning Bayesian Networks is that they can accept informative prior beliefs for each possible network, thus complementing the data.