Adding constraint support in Machine Learning has the potential to address outstanding issues in data-driven AI systems, such as safety and fairness. Existing approaches typically apply constrained optimization techniques to ML training, enforce constraint satisfaction by adjusting the model design, or use constraints to correct the output. Here, we investigate a different, complementary, strategy based on ``teaching'' constraint satisfaction to a supervised ML method via the direct use of a state-of-the-art constraint solver: this enables taking advantage of decades of research on constrained optimization with limited effort. In practice, we use a decomposition scheme alternating master steps (in charge of enforcing the constraints) and learner steps (where any supervised ML model and training algorithm can be employed). The process leads to approximate constraint satisfaction in general, and convergence properties are difficult to establish; despite this fact, we found empirically that even a na"{i}ve setup of our approach performs well on ML tasks with fairness constraints, and on classical datasets with synthetic constraints.

Teaching the Old Dog New Tricks: Supervised Learning with Constraints

Michele Lombardi
Secondo
Methodology
;
Michela Milano
Ultimo
Supervision
2021

Abstract

Adding constraint support in Machine Learning has the potential to address outstanding issues in data-driven AI systems, such as safety and fairness. Existing approaches typically apply constrained optimization techniques to ML training, enforce constraint satisfaction by adjusting the model design, or use constraints to correct the output. Here, we investigate a different, complementary, strategy based on ``teaching'' constraint satisfaction to a supervised ML method via the direct use of a state-of-the-art constraint solver: this enables taking advantage of decades of research on constrained optimization with limited effort. In practice, we use a decomposition scheme alternating master steps (in charge of enforcing the constraints) and learner steps (where any supervised ML model and training algorithm can be employed). The process leads to approximate constraint satisfaction in general, and convergence properties are difficult to establish; despite this fact, we found empirically that even a na"{i}ve setup of our approach performs well on ML tasks with fairness constraints, and on classical datasets with synthetic constraints.
Thirty-Fifth {AAAI} Conference on Artificial Intelligence, {AAAI}2021, Thirty-Third Conference on Innovative Applications of ArtificialIntelligence, {IAAI} 2021, The Eleventh Symposium on Educational Advancesin Artificial Intelligence, {EAAI} 2021, Virtual Event, February 2-9,2021
3742
3749
Fabrizio Detassis, Michele Lombardi, Michela Milano
File in questo prodotto:
File Dimensione Formato  
AAAI2021___Teaching_the_Old_Dog_new_Tricks.pdf

accesso riservato

Tipo: Versione (PDF) editoriale
Licenza: Licenza per accesso riservato
Dimensione 400.36 kB
Formato Adobe PDF
400.36 kB Adobe PDF   Visualizza/Apri   Contatta l'autore

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/861099
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? 0
social impact