DC3: A learning method for optimization with hard constraints

Large optimization problems with hard constraints arise in many settings, yet classical solvers are often prohibitively slow, motivating the use of deep networks as cheap “approximate solvers.” Unfortunately, naive deep learning approaches typically cannot enforce the hard constraints of such problems, leading to infeasible solutions. In this work, we present Deep Constraint Completion and Correction (DC3), an algorithm to address this challenge. Specifically, this method enforces feasibility via a differentiable procedure, which implicitly completes partial solutions to satisfy equality constraints and unrolls gradient-based corrections to satisfy inequality constraints. We demonstrate the effectiveness of DC3 in both synthetic optimization tasks and the real-world setting of AC optimal power flow, where hard constraints encode the physics of the electrical grid. In both cases, DC3 achieves near-optimal objective values while preserving feasibility.

[1]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

[2]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[3]  Joshua B. Tenenbaum,et al.  End-to-End Differentiable Physics for Learning and Control , 2018, NeurIPS.

[4]  Yoshua Bengio,et al.  Tackling Climate Change with Machine Learning , 2019, ACM Comput. Surv..

[5]  Gokcen Kestor,et al.  Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation , 2020, SC20: International Conference for High Performance Computing, Networking, Storage and Analysis.

[6]  Slawomir Koziel,et al.  Surrogate-Based Modeling and Optimization , 2013 .

[7]  J. Zico Kolter,et al.  What game are we playing? End-to-end learning in normal and extensive form games , 2018, IJCAI.

[8]  Yoshua Bengio,et al.  Machine Learning for Combinatorial Optimization: a Methodological Tour d'Horizon , 2018, Eur. J. Oper. Res..

[9]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[10]  Vladlen Koltun,et al.  Deep Equilibrium Models , 2019, NeurIPS.

[11]  Priya L. Donti,et al.  SATNet: Bridging deep learning and logical reasoning using a differentiable satisfiability solver , 2019, ICML.

[12]  Andreas Krause,et al.  Differentiable Learning of Submodular Models , 2017, NIPS 2017.

[13]  Max Welling,et al.  Group Equivariant Convolutional Networks , 2016, ICML.

[14]  R. Hartley,et al.  Deep Declarative Networks: A New Hope , 2019, ArXiv.

[15]  Stephen P. Boyd,et al.  OSQP: an operator splitting solver for quadratic programs , 2017, 2018 UKACC 12th International Conference on Control (CONTROL).

[16]  David Duvenaud,et al.  Neural Ordinary Differential Equations , 2018, NeurIPS.

[17]  Jason Yosinski,et al.  Hamiltonian Neural Networks , 2019, NeurIPS.

[18]  Daniel Cremers,et al.  Homogeneous Linear Inequality Constraints for Neural Network Activations , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[19]  Line Roald,et al.  Learning for Constrained Optimization: Identifying Optimal Active Constraint Sets , 2018, INFORMS J. Comput..

[20]  Amin Kargarian,et al.  A Survey on Applications of Machine Learning for Optimal Power Flow , 2020, 2020 IEEE Texas Power and Energy Conference (TPEC).

[21]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[22]  WächterAndreas,et al.  On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming , 2006 .

[23]  Michael I. Jordan,et al.  First-order methods almost always avoid saddle points: The case of vanishing step-sizes , 2019, NeurIPS.

[24]  Pierre Gentine,et al.  Achieving Conservation of Energy in Neural Network Emulators for Climate Modeling , 2019, ArXiv.

[25]  Kyri Baker,et al.  Learning Warm-Start Points For Ac Optimal Power Flow , 2019, 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP).

[26]  Kyri Baker,et al.  Learning Optimal Solutions for Extremely Fast AC Optimal Power Flow , 2019, 2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm).

[27]  Stephen P. Boyd,et al.  Differentiable Convex Optimization Layers , 2019, NeurIPS.

[28]  Andreas Krause,et al.  Differentiable Submodular Maximization , 2018, IJCAI.

[29]  Stephen P. Boyd,et al.  Solution refinement at regular points of conic problems , 2018, Computational Optimization and Applications.

[30]  Milind Tambe,et al.  Melding the Data-Decisions Pipeline: Decision-Focused Learning for Combinatorial Optimization , 2018, AAAI.

[31]  J. Zico Kolter,et al.  OptNet: Differentiable Optimization as a Layer in Neural Networks , 2017, ICML.

[32]  Priya L. Donti,et al.  Task-based End-to-end Model Learning in Stochastic Optimization , 2017, NIPS.