TY - JOUR
T1 - On the Evaluation of (Meta-)solver Approaches
AU - Amadini, Roberto
AU - Gabbrielli, Maurizio
AU - Liu, Tong
AU - Mauro, Jacopo
N1 - Publisher Copyright:
© 2023 AI Access Foundation. All rights reserved.
PY - 2023/3/17
Y1 - 2023/3/17
N2 - Meta-solver approaches exploit many individual solvers to potentially build a better solver. To assess the performance of meta-solvers, one can adopt the metrics typically used for individual solvers (e.g., runtime or solution quality) or employ more specific evaluation metrics (e.g., by measuring how close the meta-solver gets to its virtual best performance). In this paper, based on some recently published works, we provide an overview of different performance metrics for evaluating (meta-)solvers by exposing their strengths and weaknesses.
AB - Meta-solver approaches exploit many individual solvers to potentially build a better solver. To assess the performance of meta-solvers, one can adopt the metrics typically used for individual solvers (e.g., runtime or solution quality) or employ more specific evaluation metrics (e.g., by measuring how close the meta-solver gets to its virtual best performance). In this paper, based on some recently published works, we provide an overview of different performance metrics for evaluating (meta-)solvers by exposing their strengths and weaknesses.
U2 - 10.1613/JAIR.1.14102
DO - 10.1613/JAIR.1.14102
M3 - Journal article
AN - SCOPUS:85153673400
SN - 1076-9757
VL - 76
SP - 705
EP - 719
JO - Journal of Artificial Intelligence Research
JF - Journal of Artificial Intelligence Research
ER -