Optimization Proxies using Limited Labeled Data and Training Time – A Semi-Supervised Bayesian Neural Network Approach

Parikshit Pareek, Abhijith Jayakumar, Kaarthik Sundar, Sidhant Misra, Deepjyoti Deka
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:47953-47970, 2025.

Abstract

Constrained optimization problems arise in various engineering systems such as inventory management and power grids. Standard deep neural network (DNN) based machine learning proxies are ineffective in practical settings where labeled data is scarce and training times are limited. We propose a semi-supervised Bayesian Neural Networks (BNNs) based optimization proxy for this complex regime, wherein training commences in a sandwiched fashion, alternating between a supervised learning step for minimizing cost, and an unsupervised learning step for enforcing constraint feasibility. We show that the proposed semi-supervised BNN outperforms DNN architectures on important non-convex constrained optimization problems from energy network operations, achieving up to a tenfold reduction in expected maximum equality gap and halving the inequality gaps. Further, the BNN’s ability to provide posterior samples is leveraged to construct practically meaningful probabilistic confidence bounds on performance using a limited validation data, unlike prior methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-pareek25a, title = {Optimization Proxies using Limited Labeled Data and Training Time – A Semi-Supervised {B}ayesian Neural Network Approach}, author = {Pareek, Parikshit and Jayakumar, Abhijith and Sundar, Kaarthik and Misra, Sidhant and Deka, Deepjyoti}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {47953--47970}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/pareek25a/pareek25a.pdf}, url = {https://proceedings.mlr.press/v267/pareek25a.html}, abstract = {Constrained optimization problems arise in various engineering systems such as inventory management and power grids. Standard deep neural network (DNN) based machine learning proxies are ineffective in practical settings where labeled data is scarce and training times are limited. We propose a semi-supervised Bayesian Neural Networks (BNNs) based optimization proxy for this complex regime, wherein training commences in a sandwiched fashion, alternating between a supervised learning step for minimizing cost, and an unsupervised learning step for enforcing constraint feasibility. We show that the proposed semi-supervised BNN outperforms DNN architectures on important non-convex constrained optimization problems from energy network operations, achieving up to a tenfold reduction in expected maximum equality gap and halving the inequality gaps. Further, the BNN’s ability to provide posterior samples is leveraged to construct practically meaningful probabilistic confidence bounds on performance using a limited validation data, unlike prior methods.} }
Endnote
%0 Conference Paper %T Optimization Proxies using Limited Labeled Data and Training Time – A Semi-Supervised Bayesian Neural Network Approach %A Parikshit Pareek %A Abhijith Jayakumar %A Kaarthik Sundar %A Sidhant Misra %A Deepjyoti Deka %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-pareek25a %I PMLR %P 47953--47970 %U https://proceedings.mlr.press/v267/pareek25a.html %V 267 %X Constrained optimization problems arise in various engineering systems such as inventory management and power grids. Standard deep neural network (DNN) based machine learning proxies are ineffective in practical settings where labeled data is scarce and training times are limited. We propose a semi-supervised Bayesian Neural Networks (BNNs) based optimization proxy for this complex regime, wherein training commences in a sandwiched fashion, alternating between a supervised learning step for minimizing cost, and an unsupervised learning step for enforcing constraint feasibility. We show that the proposed semi-supervised BNN outperforms DNN architectures on important non-convex constrained optimization problems from energy network operations, achieving up to a tenfold reduction in expected maximum equality gap and halving the inequality gaps. Further, the BNN’s ability to provide posterior samples is leveraged to construct practically meaningful probabilistic confidence bounds on performance using a limited validation data, unlike prior methods.
APA
Pareek, P., Jayakumar, A., Sundar, K., Misra, S. & Deka, D.. (2025). Optimization Proxies using Limited Labeled Data and Training Time – A Semi-Supervised Bayesian Neural Network Approach. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:47953-47970 Available from https://proceedings.mlr.press/v267/pareek25a.html.

Related Material