Minhan Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Minhan Li has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Jayant Kalagnanam
IBM Research
2 shared publications - Katya Scheinberg
H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, USA
2 shared publications - Matt Menickelly
Argonne National Laboratory
2 shared publications - Oktay Günlük
School of Operations Research and Information Engineering, Cornell University, Ithaca, New York 14853
2 shared publications - Frank Edward Curtis
Department of Industrial and Systems Engineering, Lehigh University, Bethlehem, Pennsylvania 18015
1 shared publication - Hiva Ghanbari
Ford Motor Company, Dearborn, USA
1 shared publication
Latest Publications
- Open Access
Gradient Sampling Methods with Inexact Subproblem Solutions and Gradient Aggregation
Gradient sampling (GS) has proved to be an effective methodology for the minimization of objective functions that may be nonconvex and/or nonsmooth. The most computationally expensive component of a contemporary GS method is the need to solve a convex quadratic subproblem in each iteration. In this paper, a strategy is proposed that allows the use of inexact solutions of these subproblems, which, as proved in the paper, can be incorporated without the loss of theoretical convergence guarantees. Numerical experiments show that by exploiting inexact subproblem solutions, one can consistently reduce the computational effort required by a GS method. Additionally, a strategy is proposed for aggregating gradient information after a subproblem is solved (potentially inexactly), as has been exploited in bundle methods for nonsmooth optimization. It is proved that the aggregation scheme can be introduced without the loss of theoretical convergence guarantees. Numerical experiments show that incorporating this gradient aggregation approach substantially reduces the computational effort required by a GS method.
Open AccessGradient Sampling Methods with Inexact Subproblem Solutions and Gradient Aggregation
Gradient sampling (GS) has proved to be an effective methodology for the minimization of objective functions that may be nonconvex and/or nonsmooth. The most computationally expensive component of a contemporary GS method is the need to solve a convex quadratic subproblem in each iteration. In this paper, a strategy is proposed that allows the use of inexact solutions of these subproblems, which, as proved in the paper, can be incorporated without the loss of theoretical convergence guarantees. Numerical experiments show that by exploiting inexact subproblem solutions, one can consistently reduce the computational effort required by a GS method. Additionally, a strategy is proposed for aggregating gradient information after a subproblem is solved (potentially inexactly), as has been exploited in bundle methods for nonsmooth optimization. It is proved that the aggregation scheme can be introduced without the loss of theoretical convergence guarantees. Numerical experiments show that incorporating this gradient aggregation approach substantially reduces the computational effort required by a GS method.
- 24 March 2021
- Published by Springer Nature in Journal of Global Optimization
- Vol. 81 (1) , 233-260
Optimal decision trees for categorical data via integer programming
- 24 March 2021
- Published by Springer Nature in Journal of Global Optimization
- Vol. 81 (1) , 233-260
Optimal decision trees for categorical data via integer programming
- Open Access
Optimal Generalized Decision Trees via Integer Programming
Decision trees have been a very popular class of predictive models for decades due to their interpretability and good performance on categorical features. However, they are not always robust and tend to overfit the data. Additionally, if allowed to grow large, they lose interpretability. In this paper, we present a mixed integer programming formulation to construct optimal decision trees of a prespecified size. We take the special structure of categorical features into account and allow combinatorial decisions (based on subsets of values of features) at each node. Our approach can also handle numerical features via thresholding. We show that very good accuracy can be achieved with small trees using moderately-sized training sets. The optimization problems we solve are tractable with modern solvers.
Open AccessOptimal Generalized Decision Trees via Integer Programming
Decision trees have been a very popular class of predictive models for decades due to their interpretability and good performance on categorical features. However, they are not always robust and tend to overfit the data. Additionally, if allowed to grow large, they lose interpretability. In this paper, we present a mixed integer programming formulation to construct optimal decision trees of a prespecified size. We take the special structure of categorical features into account and allow combinatorial decisions (based on subsets of values of features) at each node. Our approach can also handle numerical features via thresholding. We show that very good accuracy can be achieved with small trees using moderately-sized training sets. The optimization problems we solve are tractable with modern solvers.
- Open Access
Novel and Efficient Approximations for Zero-One Loss of Linear Classifiers
The predictive quality of machine learning models is typically measured in terms of their (approximate) expected prediction accuracy or the so-called Area Under the Curve (AUC). Minimizing the reciprocals of these measures are the goals of supervised learning. However, when the models are constructed by the means of empirical risk minimization (ERM), surrogate functions such as the logistic loss or hinge loss are optimized instead. In this work, we show that in the case of linear predictors, the expected error and the expected ranking loss can be effectively approximated by smooth functions whose closed form expressions and those of their first (and second) order derivatives depend on the first and second moments of the data distribution, which can be precomputed. Hence, the complexity of an optimization algorithm applied to these functions does not depend on the size of the training data. These approximation functions are derived under the assumption that the output of the linear classifier for a given data set has an approximately normal distribution. We argue that this assumption is significantly weaker than the Gaussian assumption on the data itself and we support this claim by demonstrating that our new approximation is quite accurate on data sets that are not necessarily Gaussian. We present computational results that show that our proposed approximations and related optimization algorithms can produce linear classifiers with similar or better test accuracy or AUC, than those obtained using state-of-the-art approaches, in a fraction of the time.
Open AccessNovel and Efficient Approximations for Zero-One Loss of Linear Classifiers
The predictive quality of machine learning models is typically measured in terms of their (approximate) expected prediction accuracy or the so-called Area Under the Curve (AUC). Minimizing the reciprocals of these measures are the goals of supervised learning. However, when the models are constructed by the means of empirical risk minimization (ERM), surrogate functions such as the logistic loss or hinge loss are optimized instead. In this work, we show that in the case of linear predictors, the expected error and the expected ranking loss can be effectively approximated by smooth functions whose closed form expressions and those of their first (and second) order derivatives depend on the first and second moments of the data distribution, which can be precomputed. Hence, the complexity of an optimization algorithm applied to these functions does not depend on the size of the training data. These approximation functions are derived under the assumption that the output of the linear classifier for a given data set has an approximately normal distribution. We argue that this assumption is significantly weaker than the Gaussian assumption on the data itself and we support this claim by demonstrating that our new approximation is quite accurate on data sets that are not necessarily Gaussian. We present computational results that show that our proposed approximations and related optimization algorithms can produce linear classifiers with similar or better test accuracy or AUC, than those obtained using state-of-the-art approaches, in a fraction of the time.