mlsanityData to 14 Oct 2026

Cooper: A Library for Constrained Optimization in Deep Learning

arXiv 2504.01212 ↗First posted 1 Apr 2025cs.LG cs.MScited by 11

Not on the map. Why?

Abstract (arXiv)

Cooper is an open-source package for solving constrained optimization problems involving deep learning models. Cooper implements several Lagrangian-based first-order update schemes, making it easy to combine constrained optimization algorithms with high-level features of PyTorch such as automatic differentiation, and specialized deep learning architectures and optimizers. Although Cooper is specifically designed for deep learning applications where gradients are estimated based on mini-batches, it is suitable for general non-convex continuous constrained optimization. Cooper's source code is available at https://github.com/cooper-org/cooper.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 21 of its 29 references that are in the map. First posted on arXiv 1 Apr 2025.

Builds on

21 of 29 references.

References · 21

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2017Deep reinforcement learning from human preferencescited by 3,912
  4. 2021Convex optimizationcited by 2,515
  5. 2023Safe RLHF: Safe Reinforcement Learning from Human Feedbackcited by 656
  6. 2016CVXPY: A Python-Embedded Modeling Language for Convex Optimizationcited by 641
  7. 2020Responsive Safety in Reinforcement Learning by PID Lagrangian Methodscited by 316
  8. 2021Efficient and Modular Implicit Differentiationcited by 203
  9. 2018A Variational Inequality Perspective on Generative Adversarial Networkscited by 144
  10. 2018Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goalscited by 103
  11. 2024Nonparametric Partial Disentanglement via Mechanism Sparsity: Sparse Actions, Interventions and Sparse Temporal Dependenciescited by 41
  12. 2023Generalized Logit Adjustment: Calibrating Fine-tuned Models by Removing Label Bias in Foundation Modelscited by 32
9 more
  1. 2022Partial Disentanglement via Mechanism Sparsitycited by 29
  2. 2022Controlled Sparsity via Constrained Optimization or: How I Learned to Stop Tuning Penalties and Love Constraintscited by 25
  3. 2022A Lagrangian Duality Approach to Active Learningcited by 24
  4. 2024On PI Controllers for Updating Lagrange Multipliers in Constrained Optimizationcited by 12
  5. 2023Balancing Act: Constraining Disparate Impact in Sparse Modelscited by 7
  6. 2022L0onie: Compressing COINs with L0-constraintscited by 4
  7. 2022Neural Networks with Quantization Constraintscited by 4
  8. 2023Active Reinforcement Learning for Robust Building Controlcited by 4
  9. 2024Novel Node Category Detection Under Subpopulation Shiftcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 6.

11 papers.

  1. 2026Audit Me If You Can: Query-Efficient Active Fairness Auditing of Black-Box LLMscited by 6
  2. 2026Stop Probing, Start Coding: Why Linear Probes and Sparse Autoencoders Fail at Compositional Generalisationcited by 6
  3. 2025Position: Adopt Constraints Over Fixed Penalties in Deep Learningcited by 5
  4. 2025Provably Safe Model Updatescited by 3
  5. 2025humancompatible.train: Implementing Optimization Algorithms for Stochastically-Constrained Stochastic Optimization Problemscited by 1
  6. 2025Dual Optimistic Ascent (PI Control) is the Augmented Lagrangian Method in Disguisecited by 1
  7. 2026Unsupervised Causal Abstractions Discoverycited by 1
  8. 2026Not All Objectives Are Born Equal: Priority-Constrained Descent for Hierarchical Multi-Objective Optimizationcited by 1
  9. 2025Benchmarking Stochastic Approximation Algorithms for Fairness-Constrained Training of Deep Neural Networkscited by 0
  10. 2026Stochastic Penalty-Barrier Method for Constrained Machine Learningcited by 0
  11. 2026Neural Slack Variables for Shape Constraintscited by 0