Solving Functional Optimization with Deep Networks and Variational Principles

• URL: http://arxiv.org/abs/2410.06277v4
• Date: Tue, 11 Mar 2025 21:28:20 GMT
• Title: Solving Functional Optimization with Deep Networks and Variational Principles
• Authors: Kawisorn Kamtue, Jose M. F. Moura, Orathai Sangpetch,
• Abstract summary: We show how to leverage the calculus of variations to design deep neural networks to solve functional optimization problems.<n>Our approach is crucial when the solution is a function defined over an unknown interval or supporttextemdash such as in minimum-time control problems.<n>We show that CalVNet can be trained in an unsupervised manner, without relying on ground-truth data.
• Score: 0.2633434651741688
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Can neural networks solve math problems using first a principle alone? This paper shows how to leverage the fundamental theorem of the calculus of variations to design deep neural networks to solve functional optimization without requiring training data (e.g., ground-truth optimal solutions). Our approach is particularly crucial when the solution is a function defined over an unknown interval or support\textemdash such as in minimum-time control problems. By incorporating the necessary conditions satisfied by the optimal function solution, as derived from the calculus of variation, in the design of the deep architecture, CalVNet leverages overparameterized neural networks to learn these optimal functions directly. We validate CalVNet by showing that, without relying on ground-truth data and simply incorporating first principles, it successfully derives the Kalman filter for linear filtering, the bang-bang optimal control for minimum-time problems, and finds geodesics on manifolds. Our results demonstrate that CalVNet can be trained in an unsupervised manner, without relying on ground-truth data, establishing a promising framework for addressing general, potentially unsolved functional optimization problems that still lack analytical solutions.

Related papers

• A Neural Network Framework for Discovering Closed-form Solutions to Quadratic Programs with Linear Constraints [2.064612766965483]
  Deep neural networks (DNNs) have been used to model complex optimization problems in many applications.<n>They have difficulty guaranteeing solution optimality and feasibility, despite training on large datasets.<n>This paper proposes a NN modeling approach and learning algorithm that discovers the exact closed-form solution to a quadratic program with linear constraints.
  arXiv  Detail & Related papers  (2025-10-27T18:11:44Z)
• Were Residual Penalty and Neural Operators All We Needed for Solving Optimal Control Problems? [0.0]
  We show that a simple neural operator architecture, such as DeepONet, can solve multiple optimal control problems with a single physics-informed training phase and a subsequent optimization phase.<n>Our results show acceptable accuracy for practical applications and potential computational savings for more complex and higher-dimensional problems.
  arXiv  Detail & Related papers  (2025-06-05T08:22:16Z)
• Learning Multiple Initial Solutions to Optimization Problems [52.9380464408756]
  Sequentially solving similar optimization problems under strict runtime constraints is essential for many applications. We propose learning to predict emphmultiple diverse initial solutions given parameters that define the problem instance. We find significant and consistent improvement with our method across all evaluation settings and demonstrate that it efficiently scales with the number of initial solutions required.
  arXiv  Detail & Related papers  (2024-11-04T15:17:19Z)
• Self-Supervised Learning of Iterative Solvers for Constrained Optimization [0.0]
  We propose a learning-based iterative solver for constrained optimization. It can obtain very fast and accurate solutions by customizing the solver to a specific parametric optimization problem. A novel loss function based on the Karush-Kuhn-Tucker conditions of optimality is introduced, enabling fully self-supervised training of both neural networks.
  arXiv  Detail & Related papers  (2024-09-12T14:17:23Z)
• The Unreasonable Effectiveness of Solving Inverse Problems with Neural Networks [24.766470360665647]
  We show that neural networks trained to learn solutions to inverse problems can find better solutions than classicals even on their training set. Our findings suggest an alternative use for neural networks: rather than generalizing to new data for fast inference, they can also be used to find better solutions on known data.
  arXiv  Detail & Related papers  (2024-08-15T12:38:10Z)
• Neural Time-Reversed Generalized Riccati Equation [60.92253836775246]
  Hamiltonian equations offer an interpretation of optimality through auxiliary variables known as costates. This paper introduces a novel neural-based approach to optimal control, with the aim of working forward-in-time.
  arXiv  Detail & Related papers  (2023-12-14T19:29:37Z)
• Stochastic Optimal Control Matching [53.156277491861985]
  Our work introduces Optimal Control Matching (SOCM), a novel Iterative Diffusion Optimization (IDO) technique for optimal control. The control is learned via a least squares problem by trying to fit a matching vector field. Experimentally, our algorithm achieves lower error than all the existing IDO techniques for optimal control.
  arXiv  Detail & Related papers  (2023-12-04T16:49:43Z)
• Safe Neural Control for Non-Affine Control Systems with Differentiable Control Barrier Functions [58.19198103790931]
  This paper addresses the problem of safety-critical control for non-affine control systems. It has been shown that optimizing quadratic costs subject to state and control constraints can be sub-optimally reduced to a sequence of quadratic programs (QPs) by using Control Barrier Functions (CBFs) We incorporate higher-order CBFs into neural ordinary differential equation-based learning models as differentiable CBFs to guarantee safety for non-affine control systems.
  arXiv  Detail & Related papers  (2023-09-06T05:35:48Z)
• An Application of Pontryagin Neural Networks to Solve Optimal Quantum Control Problems [1.5469452301122175]
  Pontryagin maximum principle has proved to play an important role to achieve the maximum fidelity in an optimum time or energy. We formulate a control constrained optimal control problem where we aim to minimize time and also energy subjected to a quantum system satisfying the bilinear Schrodinger equation. We make use of the so-called "qutip" package in python, and the newly developed "tfc" python package.
  arXiv  Detail & Related papers  (2023-02-01T17:48:07Z)
• Pontryagin Optimal Control via Neural Networks [19.546571122359534]
  We integrate Neural Networks with the Pontryagin's Maximum Principle (PMP), and propose a sample efficient framework NN-PMP-Gradient. The resulting controller can be implemented for systems with unknown and complex dynamics. Compared with the widely applied model-free and model-based reinforcement learning (RL) algorithms, our NN-PMP-Gradient achieves higher sample-efficiency and performance in terms of control objectives.
  arXiv  Detail & Related papers  (2022-12-30T06:47:03Z)
• Unsupervised Optimal Power Flow Using Graph Neural Networks [172.33624307594158]
  We use a graph neural network to learn a nonlinear parametrization between the power demanded and the corresponding allocation. We show through simulations that the use of GNNs in this unsupervised learning context leads to solutions comparable to standard solvers.
  arXiv  Detail & Related papers  (2022-10-17T17:30:09Z)
• Optimization-Informed Neural Networks [0.6853165736531939]
  We propose optimization-informed neural networks (OINN) to solve constrained nonlinear optimization problems. In a nutshell, OINN transforms a CNLP into a neural network training problem. The effectiveness of the proposed approach is demonstrated through a collection of classical problems.
  arXiv  Detail & Related papers  (2022-10-05T09:28:55Z)
• Application of Pontryagin's Maximum Principle to Quantum Metrology in Dissipative Systems [8.920103626492315]
  We look for the optimal control that maximizes quantum Fisher information for "twist and turn" problem. We find that the optimal control is singular without dissipation but can become unbounded once the quantum decoherence is introduced.
  arXiv  Detail & Related papers  (2022-04-30T00:02:57Z)
• Adaptive neural domain refinement for solving time-dependent differential equations [0.0]
  A classic approach for solving differential equations with neural networks builds upon neural forms, which employ the differential equation with a discretisation of the solution domain. It would be desirable to transfer such important and successful strategies to the field of neural network based solutions. We propose a novel adaptive neural approach to meet this aim for solving time-dependent problems.
  arXiv  Detail & Related papers  (2021-12-23T13:19:07Z)
• Path Regularization: A Convexity and Sparsity Inducing Regularization for Parallel ReLU Networks [75.33431791218302]
  We study the training problem of deep neural networks and introduce an analytic approach to unveil hidden convexity in the optimization landscape. We consider a deep parallel ReLU network architecture, which also includes standard deep networks and ResNets as its special cases.
  arXiv  Detail & Related papers  (2021-10-18T18:00:36Z)
• Neural Predictive Control for the Optimization of Smart Grid Flexibility Schedules [0.0]
  Model predictive control (MPC) is a method to formulate the optimal scheduling problem for grid flexibilities in a mathematical manner. MPC methods promise accurate results for time-constrained grid optimization but they are inherently limited by the calculation time needed for large and complex power system models. A Neural Predictive Control scheme is proposed to learn optimal control policies for linear and nonlinear power systems through imitation.
  arXiv  Detail & Related papers  (2021-08-19T15:12:35Z)
• Offline Model-Based Optimization via Normalized Maximum Likelihood Estimation [101.22379613810881]
  We consider data-driven optimization problems where one must maximize a function given only queries at a fixed set of points. This problem setting emerges in many domains where function evaluation is a complex and expensive process. We propose a tractable approximation that allows us to scale our method to high-capacity neural network models.
  arXiv  Detail & Related papers  (2021-02-16T06:04:27Z)
• Unsupervised Deep Learning for Optimizing Wireless Systems with Instantaneous and Statistic Constraints [29.823814915538463]
  We establish a unified framework of using unsupervised deep learning to solve both kinds of problems with both instantaneous and statistic constraints. We show that unsupervised learning outperforms supervised learning in terms of violation probability and approximation accuracy of the optimal policy.
  arXiv  Detail & Related papers  (2020-05-30T13:37:14Z)
• Physarum Powered Differentiable Linear Programming Layers and Applications [48.77235931652611]
  We propose an efficient and differentiable solver for general linear programming problems. We show the use of our solver in a video segmentation task and meta-learning for few-shot learning.
  arXiv  Detail & Related papers  (2020-04-30T01:50:37Z)
• Self-Directed Online Machine Learning for Topology Optimization [58.920693413667216]
  Self-directed Online Learning Optimization integrates Deep Neural Network (DNN) with Finite Element Method (FEM) calculations. Our algorithm was tested by four types of problems including compliance minimization, fluid-structure optimization, heat transfer enhancement and truss optimization. It reduced the computational time by 2 5 orders of magnitude compared with directly using methods, and outperformed all state-of-the-art algorithms tested in our experiments.
  arXiv  Detail & Related papers  (2020-02-04T20:00:28Z)
• Optimizing Wireless Systems Using Unsupervised and Reinforced-Unsupervised Deep Learning [96.01176486957226]
  Resource allocation and transceivers in wireless networks are usually designed by solving optimization problems. In this article, we introduce unsupervised and reinforced-unsupervised learning frameworks for solving both variable and functional optimization problems.
  arXiv  Detail & Related papers  (2020-01-03T11:01:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.