Q-Intersection Algorithms for Constraint-Based Robust Parameter Estimation

Authors: Clement Carbonnel, Gilles Trombettoni, Philippe Vismara, Gilles Chabert

AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present a computational study of the q-intersection. We also propose a fast heuristic and a sophisticated exact q-intersection algorithm. First experiments show that our exact algorithm outperforms the existing one while our heuristic performs an efficient filtering on hard problems.
Researcher Affiliation Academia Clement Carbonnel LAAS-CNRS Universit e Toulouse, France carbonnel@laas.fr Gilles Trombettoni LIRMM Universit e Montpellier, France gilles.trombettoni@lirmm.fr Philippe Vismara LIRMM Sup Agro Montpellier, France vismara@lirmm.fr Gilles Chabert LINA EMN Nantes, France gilles.chabert@mines-nantes.fr
Pseudocode Yes Algorithm 1: core F(S, q); Algorithm 2: QInter2(S, q); Algorithm 3: Find Min Q(S, q, d, N)
Open Source Code Yes The code and instances of our experiments are available in the release 2.1.2 (and subsequent) of IBEX (www.ibex-lib.org).
Open Datasets No The paper uses "randomly generated instances" for its experiments, as stated in Section 5. It does not refer to a publicly available or open dataset with a specific link or citation.
Dataset Splits No The paper describes generating instances and comparing algorithms, but it does not specify training, validation, or test dataset splits. The experimental setup is not structured like a typical machine learning model evaluation with explicit data splits for these purposes.
Hardware Specification Yes All experiments have been performed on a 2.2Ghz Intel Core i7.
Software Dependencies Yes We have implemented the projective filtering, core F and QInter2 in the C++ numerical constraint programming library Ibex (Chabert and Jaulin 2009; Chabert 2014), which originally shipped with the grid algorithm. Q-clique problems are solved using Cliquer (Ostergard 2002). The code and instances of our experiments are available in the release 2.1.2 (and subsequent) of IBEX (www.ibex-lib.org).
Experiment Setup Yes The center of each of the p boxes in S and its radius in every dimension are uniformly drawn in [0, L]. The two values of qc we have studied are 0.3 p (many outliers) and 0.8 p (few outliers). The corresponding scaling factor (for transforming each random instance into a difficult one) is determined empirically.