Skip to main navigation Skip to main content

This methodology thrust investigates three research avenues:

  • Optimization Proxies: the idea of replacing a time-consuming optimization model by a machine-learning proxy that can be used in real time and/or in computational demanding applications (e.g., large-scale Monte-Carlo simulations). The challenge is to learn such proxies for large-scale optimization problems that include physical, engineering, and operational constraints. The proxies can also be improved by reinforcement learning when they are part of as sequential decision process.
  • End-to-End Learning: the idea of integration of optimization layers as parts of the deep-learning pipeline. The challenge is to define combinatorial layers capturing specific families of combinatorial optimization tasks that are amenable to useful differentiable optimization surrogates.
  • Learning to optimize: the idea of replacing components of optimization models by machine-learning models. 

Here are a teaser and a longer presentation about optimization proxies with applications in energy, supply chains, and mobility.

  

The Institute is pioneering a number of new methodological directions in this space:

These methodologies have been applied to real-case studies in power systems and transportation systems.

UCLA IPAM Presentation on Online and Offine Learning for Contextual Stochastic Learning

Publications

Funded by the Institute

Related work

  • Reinforcement Learning from Optimization Proxy for Ride-Hailing Vehicle Relocation. Enpeng Yuan, Wenbo Chen, and Pascal Van Hentenryck. Journal of Artificial Intelligence Research (JAIR),  75, 985-1002, November, 2022.

  • Spatial Network Decomposition for Fast and Scalable AC-OPF Learning.Minas Chatzos, Terrence Mak,  and Pascal Van Hentenryck. IEEE Transactions on Power Systems, 37(4), 2601--2612, July 2022.

  • Spatial Network Decomposition for Fast and Scalable AC-OPF Learning. Minas Chatzos, Terrence Mak, and Pascal Van Hentenryck. IEEE Transactions on Power Systems (Early access: 10.1109/TPWRS.2021.3124726).

  • Machine Learning for Optimal Power Flows.  Pascal Van Hentenryck. INFORMS TutORials 2021.

  • End- to-End Constrained Optimization Learning: A Survey. James Kotary, Ferdinando Fioretto, Pascal Van Hentenryck, and Bryan Wilder. In the 30th International Joint Conference on Artificial Intelligence (IJCAI-21), Montreal, Canada, August, 2021.

  • Combining Deep Learning and Optimization for Preventive Security-Constrained DC Optimal PowerFlow.Alexandre Velloso and Pascal Van Hentenryck. IEEE Transactions on Power Systems, 36(4), July 2021.

  • Real-time dispatching of large-scale ride-sharing systems: Integrating optimization, machine learning, and model predictive control. Connor Riley, Pascal Van Hentenryck, and Enpeng YuanIn Christian Bessiere, editor, Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI 2020, pages 4417–4423. ijcai.org, 2020.

  • Lagrangian Duality for Constrained Deep Learning. Ferdinando Fioretto, Pascal Van Hentenryck, Terrence WK Mak, Cuong Tran, Federico Baldo, and Michele Lombardi. In Proceedings of 2020 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, June 2020.

  • Mipaal: Mixed integer program as a layer. Aaron Ferber, Bryan Wilder, Bistra Dilkina, and Milind Tambe. In AAAI Conference on Artificial Intelligence, pages 1504–1511, 2020.

  • Predicting AC optimal power flows: Combining deep learning and lagrangian dual methods. Ferdinando Fioretto, Terrence W.K. Mak, and Pascal Van Hentenryck. Proceedings of the AAAI Conference on Artificial Intelligence, 34(01):630–637,  2020.

  • Real-time dispatching of large-scale ride-sharing systems: Integrating optimization, machine learning, and model predictive control. Connor Riley, Pascal Van Hentenryck, and Enpeng Yuan. In Christian Bessiere, editor, Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, IJCAI 2020, pages 4417–4423. 

  • End-to-end game-focused learning of adversary behavior in security games.  Andrew Perrault, Bryan Wilder, Eric Ewing, Aditya Mate, Bistra Dilkina, and Milind Tambe. In AAAI Conference on Artificial Intelligence, pages 1378–1386, 2020.

  • Melding the Data-Decisions Pipeline: Decision-Focused Learning for Combinatorial Optimization. B. Wilder, B. Dilkina, and M. Tambe. In AAAI Conference on Artificial Intelligence, 2019.

  • Bryan Wilder, Eric Ewing, Bistra Dilkina, and Milind Tambe. End to end learning and optimization on graphs. In Advances in Neural Information Processing Systems, pages 4672–4683, 2019.

End to End Learning and Optimization Team

Pascal Van Hentenryck
Pascal Van Hentenryck
Director
Georgia Institute of Technology
Leader
Bistra Dilkina
Bistra Dilkina
USC Site Director
University of Southern California
Co-leader
Santanu Dey
Santanu Dey
Georgia Institute of Technology
Paul Grigas
Paul Grigas
University of California, Berkeley
Barna Saha
Barna Saha
UCSD Site
University of California, San Diego
Tuo Zhao
Tuo Zhao
Georgia Institute of Technology