Addressing the Meta-Learning problem with Metaheuristics

William Raynaut (1, 2) , Chantal Soulé-Dupuy (1, 3) , Nathalie Vallès-Parlangeau (1, 3) , Cédric Dray (4) , Philippe Valet (4)

1 IRIT-SIG - Systèmes d’Informations Généralisées
2 UT3 - Université Toulouse III - Paul Sabatier
3 UT Capitole - Université Toulouse Capitole
4 I2MC - Institut des Maladies Métaboliques et Cardiovasculaires

The many machine learning and data mining techniques produced over the last decades can prove invaluable assets in diverse fields, but choosing the most appropriate for a given application may be very difficult for a non-expert. Our objective is thus to provide modelling assistance using a meta-learning approach based on an evolutionary metaheuristic. We present the intended workflow of such modelling assistant and the expected challenges along our line of work.

Domaines: Intelligence artificielle [cs.AI]; Apprentissage [cs.LG]

William Raynaut, Chantal Soulé-Dupuy, Nathalie Vallès-Parlangeau, Cédric Dray, Philippe Valet. Addressing the Meta-Learning problem with Metaheuristics. 11th Metaheuristics International Conference (MIC 2015), Jun 2015, Agadir, Morocco. pp.91-92. ⟨hal-01470871⟩