TY - JOUR
T1 - Learning Customised Decision Trees for Domain-knowledge Constraints
AU - Nanfack, Géraldin
AU - Temple, Paul
AU - Frénay, Benoît
N1 - Publisher Copyright:
© 2023
PY - 2023/10
Y1 - 2023/10
N2 - When applied to critical domains, machine learning models usually need to comply with prior knowledge and domain-specific requirements. For example, one may require that a learned decision tree model should be of limited size and fair, so as to be easily interpretable, trusted, and adopted. However, most state-of-the-art models, even on decision trees, only aim to maximising expected accuracy. In this paper, we propose a framework in which a diverse family of prior and domain knowledge can be formalised and imposed as constraints on decision trees. This framework is built upon a newly introduced tree representation that leads to two generic linear programming formulations of the optimal decision tree problem. The first one targets binary features, while the second one handles continuous features without the need for discretisation. We theoretically show how a diverse family of constraints can be formalised in our framework. We validate the framework with constraints on several applications and perform extensive experiments, demonstrating empirical evidence of comparable performance w.r.t. state-of-the-art tree learners.
AB - When applied to critical domains, machine learning models usually need to comply with prior knowledge and domain-specific requirements. For example, one may require that a learned decision tree model should be of limited size and fair, so as to be easily interpretable, trusted, and adopted. However, most state-of-the-art models, even on decision trees, only aim to maximising expected accuracy. In this paper, we propose a framework in which a diverse family of prior and domain knowledge can be formalised and imposed as constraints on decision trees. This framework is built upon a newly introduced tree representation that leads to two generic linear programming formulations of the optimal decision tree problem. The first one targets binary features, while the second one handles continuous features without the need for discretisation. We theoretically show how a diverse family of constraints can be formalised in our framework. We validate the framework with constraints on several applications and perform extensive experiments, demonstrating empirical evidence of comparable performance w.r.t. state-of-the-art tree learners.
KW - Constraints
KW - Decision trees
KW - Domain knowledge
UR - http://www.scopus.com/inward/record.url?scp=85160789765&partnerID=8YFLogxK
U2 - 10.1016/j.patcog.2023.109610
DO - 10.1016/j.patcog.2023.109610
M3 - Article
AN - SCOPUS:85160789765
SN - 0031-3203
VL - 142
JO - Pattern Recognition
JF - Pattern Recognition
M1 - 109610
ER -