Minimizing Surrogate Losses for Decision-Focused Learning using Differentiable Optimization

• URL: http://arxiv.org/abs/2508.11365v2
• Date: Mon, 25 Aug 2025 12:45:34 GMT
• Title: Minimizing Surrogate Losses for Decision-Focused Learning using Differentiable Optimization
• Abstract summary: Decision-focused learning (DFL) trains a machine learning (ML) model to predict parameters of an optimization problem.<n> Gradient-based DFL requires computing the derivative of the solution to the optimization problem with respect to the predicted parameters.<n>For many optimization problems, such as linear programs (LPs), the gradient of the regret with respect to the predicted parameters is zero almost everywhere.
• Score: 7.597423277859608
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Decision-focused learning (DFL) trains a machine learning (ML) model to predict parameters of an optimization problem, to directly minimize decision regret, i.e., maximize decision quality. Gradient-based DFL requires computing the derivative of the solution to the optimization problem with respect to the predicted parameters. However, for many optimization problems, such as linear programs (LPs), the gradient of the regret with respect to the predicted parameters is zero almost everywhere. Existing gradient-based DFL approaches for LPs try to circumvent this issue in one of two ways: (a) smoothing the LP into a differentiable optimization problem by adding a quadratic regularizer and then minimizing the regret directly or (b) minimizing surrogate losses that have informative (sub)gradients. In this paper, we show that the former approach still results in zero gradients, because even after smoothing the regret remains constant across large regions of the parameter space. To address this, we propose minimizing surrogate losses -- even when a differentiable optimization layer is used and regret can be minimized directly. Our experiments demonstrate that minimizing surrogate losses allows differentiable optimization layers to achieve regret comparable to or better than surrogate-loss based DFL methods. Further, we demonstrate that this also holds for DYS-Net, a recently proposed differentiable optimization technique for LPs, that computes approximate solutions and gradients through operations that can be performed using feedforward neural network layers. Because DYS-Net executes the forward and the backward pass very efficiently, by minimizing surrogate losses using DYS-Net, we are able to attain regret on par with the state-of-the-art while reducing training time by a significant margin.

Related papers

• Decision-Focused Learning via Tangent-Space Projection of Prediction Error [32.655683009031804]
  Decision-Focused Learning (DFL) trains predictors to improve downstream decision quality.<n>We show that, under standard regularity with locally stable active constraints, the regret gradient admits a closed-form geometric characterization.<n>We propose PEAR, which computes regret gradients via a reduced linear system over active constraints.
  arXiv  Detail & Related papers  (2026-05-02T10:10:16Z)
• Optimizing Optimizers for Fast Gradient-Based Learning [53.81268610971847]
  We lay the theoretical foundation for automating design in gradient-based learning.<n>By treating gradient loss signals as a function that translates to parameter motions, the problem reduces to a family of convex optimization problems.
  arXiv  Detail & Related papers  (2025-12-06T09:50:41Z)
• Solver-Free Decision-Focused Learning for Linear Optimization Problems [6.305123652677644]
  In many real-world scenarios, the parameters of the optimization problem are not known a priori and must be predicted from contextual features.<n>This gives rise to predict-then-optimize problems, where a machine learning model predicts problem parameters that are then used to make decisions via optimization.<n>We propose a solver-free training method that exploits the geometric structure of linear optimization to enable efficient training with minimal degradation in solution quality.
  arXiv  Detail & Related papers  (2025-05-28T10:55:16Z)
• AYLA: Amplifying Gradient Sensitivity via Loss Transformation in Non-Convex Optimization [0.0]
  Gradient Descent (SGD) and its variants, such as ADAM, are foundational to deep learning optimization.<n>This paper introduces AYLA, a novel framework that enhances dynamics training.
  arXiv  Detail & Related papers  (2025-04-02T16:31:39Z)
• Training Deep Learning Models with Norm-Constrained LMOs [56.00317694850397]
  We propose a new family of algorithms that uses the linear minimization oracle (LMO) to adapt to the geometry of the problem.<n>We demonstrate significant speedups on nanoGPT training using our algorithm, Scion, without any reliance on Adam.
  arXiv  Detail & Related papers  (2025-02-11T13:10:34Z)
• Zeroth-Order Fine-Tuning of LLMs in Random Subspaces [63.10833446782114]
  As language models grow in size, memory demands for backpropagation increase.<n>Zeroth-order (ZO) optimization methods offer a memory-efficient alternative.<n>In this paper, we propose Subspace Zero-order optimization to address the challenges posed by posed by high dimensionality perturbations.
  arXiv  Detail & Related papers  (2024-10-11T17:01:43Z)
• Gradient-free neural topology optimization [0.0]
  gradient-free algorithms require many more iterations to converge when compared to gradient-based algorithms. This has made them unviable for topology optimization due to the high computational cost per iteration and high dimensionality of these problems. We propose a pre-trained neural reparameterization strategy that leads to at least one order of magnitude decrease in iteration count when optimizing the designs in latent space.
  arXiv  Detail & Related papers  (2024-03-07T23:00:49Z)
• Multi-fidelity Constrained Optimization for Stochastic Black Box Simulators [1.6385815610837167]
  We introduce the algorithm Scout-Nd (Stochastic Constrained Optimization for N dimensions) to tackle the issues mentioned earlier. Scout-Nd efficiently estimates the gradient, reduces the noise of the estimator gradient, and applies multi-fidelity schemes to further reduce computational effort. We validate our approach on standard benchmarks, demonstrating its effectiveness in optimizing parameters highlighting better performance compared to existing methods.
  arXiv  Detail & Related papers  (2023-11-25T23:36:38Z)
• Score Function Gradient Estimation to Widen the Applicability of Decision-Focused Learning [17.962860438133312]
  Decision-focused learning (DFL) paradigm overcomes limitation by training to directly minimize a task loss, e.g. regret. We propose an alternative method that makes no such assumptions, it combines smoothing with score function estimation which works on any task loss. Experiments show that it typically requires more epochs, but that it is on par with specialized methods and performs especially well for the difficult case of problems with uncertainty in the constraints, in terms of solution quality, scalability, or both.
  arXiv  Detail & Related papers  (2023-07-11T12:32:13Z)
• Scaling Forward Gradient With Local Losses [117.22685584919756]
  Forward learning is a biologically plausible alternative to backprop for learning deep neural networks. We show that it is possible to substantially reduce the variance of the forward gradient by applying perturbations to activations rather than weights. Our approach matches backprop on MNIST and CIFAR-10 and significantly outperforms previously proposed backprop-free algorithms on ImageNet.
  arXiv  Detail & Related papers  (2022-10-07T03:52:27Z)
• Second-Order Sensitivity Analysis for Bilevel Optimization [26.17470185675129]
  We extend sensitivity analysis to provide second-order derivative information of the lower problem. We show that much of the computation already used to produce the IFT gradient can be reused for the IFT Hessian.
  arXiv  Detail & Related papers  (2022-05-04T21:16:05Z)
• Implicit Parameter-free Online Learning with Truncated Linear Models [51.71216912089413]
  parameter-free algorithms are online learning algorithms that do not require setting learning rates. We propose new parameter-free algorithms that can take advantage of truncated linear models through a new update that has an "implicit" flavor. Based on a novel decomposition of the regret, the new update is efficient, requires only one gradient at each step, never overshoots the minimum of the truncated model, and retains the favorable parameter-free properties.
  arXiv  Detail & Related papers  (2022-03-19T13:39:49Z)
• Extrapolation for Large-batch Training in Deep Learning [72.61259487233214]
  We show that a host of variations can be covered in a unified framework that we propose. We prove the convergence of this novel scheme and rigorously evaluate its empirical performance on ResNet, LSTM, and Transformer.
  arXiv  Detail & Related papers  (2020-06-10T08:22:41Z)

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.