mlsanityData to 7 Oct 2026

Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds

arXiv 1804.05345 ↗First posted 15 Apr 2018cs.LG cs.DS stat.MLcited by 43

Not on the map. Why?

Abstract (arXiv)

We present an efficient coresets-based neural network compression algorithm that sparsifies the parameters of a trained fully-connected neural network in a manner that provably approximates the network's output. Our approach is based on an importance sampling scheme that judiciously defines a sampling distribution over the neural network parameters, and as a result, retains parameters of high importance while discarding redundant ones. We leverage a novel, empirical notion of sensitivity and extend traditional coreset constructions to the application of compressing parameters. Our theoretical analysis establishes guarantees on the size and accuracy of the resulting compressed network and gives rise to generalization bounds that may provide new insights into the generalization properties of neural networks. We demonstrate the practical effectiveness of our algorithm on a variety of neural network configurations and real-world data sets.

Key facts. Cited by 43 papers in the map (data through 7 Oct 2026). First posted on arXiv 15 Apr 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 12; 2022: 13; 2023: 8; 2024: 1; 2025: 3; 2026: 6.

43 papers.

Sort:

By area:

  1. 2021Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networkscited by 521
  2. 2023Structured Pruning for Deep Convolutional Neural Networks: A surveycited by 122
  3. 2021Lost in Pruning: The Effects of Pruning Neural Networks beyond Test Accuracycited by 62
  4. 2022The Combinatorial Brain Surgeon: Pruning Weights That Cancel One Another in Neural Networkscited by 41
  5. 2021Compressing Neural Networks: Towards Determining the Optimal Layer-wise Decompositioncited by 33
  6. 2022Pruning's Effect on Generalization Through the Lens of Training and Regularizationcited by 31
  7. 2023Gradient-Free Structured Pruning with Unlabeled Datacited by 26
  8. 2023Pruning Deep Neural Networks from a Sparsity Perspectivecited by 25
  9. 2022Rate-Distortion Theoretic Generalization Bounds for Stochastic Learning Algorithmscited by 22
  10. 2022Pruning Neural Networks via Coresets and Convex Geometry: Towards No Assumptionscited by 20
  11. 2021A Unified Approach to Coreset Learningcited by 18
  12. 2021A Probabilistic Approach to Neural Network Pruningcited by 15
31 more
  1. 2021Why Lottery Ticket Wins? A Theoretical Perspective of Sample Complexity on Pruned Neural Networkscited by 14
  2. 2021Sparse Flows: Pruning Continuous-depth Modelscited by 13
  3. 2021Adversarial Robustness of Streaming Algorithms through Importance Samplingcited by 13
  4. 2022On Measuring Excess Capacity in Neural Networkscited by 9
  5. 2022Recall Distortion in Neural Network Pruning and the Undecayed Pruning Algorithmcited by 9
  6. 2023Getting Away with More Network Pruning: From Sparsity to Geometry and Linear Regionscited by 8
  7. 2022Coresets for Data Discretization and Sine Wave Fittingcited by 6
  8. 2022Obstacle Aware Sampling for Path Planningcited by 6
  9. 2023Data-Efficient Training of CNNs and Transformers with Coresets: A Stability Perspectivecited by 6
  10. 2022A Theoretical Understanding of Neural Network Compression from Sparse Linear Approximationcited by 5
  11. 2021Low-Regret Active learningcited by 3
  12. 2022Robust Learning of Parsimonious Deep Neural Networkscited by 3
  13. 2021A Deeper Look into Convolutions via Eigenvalue-based Pruningcited by 2
  14. 2021Probabilistic fine-tuning of pruning masks and PAC-Bayes self-bounded learningcited by 2
  15. 2022The Effect of Data Dimensionality on Neural Network Prunabilitycited by 2
  16. 2023How Informative is the Approximation Error from Tensor Decomposition for Neural Network Compression?cited by 2
  17. 2024"Forgetting" in Machine Learning and Beyond: A Surveycited by 2
  18. 2021i-SpaSP: Structured Neural Pruning via Sparse Signal Recoverycited by 1
  19. 2025Compress to Impress: Efficient LLM Adaptation Using a Single Gradient Step on 100 Samplescited by 1
  20. 2025ModHiFi: Identifying High Fidelity predictive components for Model Modificationcited by 1
  21. 2025Theoretical Compression Bounds for Wide Multilayer Perceptronscited by 1
  22. 2026Provable Pruning for Efficient 3D Gaussian Splatting via Coresetscited by 1
  23. 2022Approximability and Generalisationcited by 0
  24. 2022Efficient NTK using Dimensionality Reductioncited by 0
  25. 2023Non-Convex Optimizations for Machine Learning with Theoretical Guarantee: Robust Matrix Completion and Neural Network Learningcited by 0
  26. 2023Integrating Fairness and Model Pruning Through Bi-level Optimizationcited by 0
  27. 2026Deterministic Coreset for Lp Subspacecited by 0
  28. 2026FedProxy: Federated Fine-Tuning of LLMs via Proxy SLMs and Heterogeneity-Aware Fusioncited by 0
  29. 2026Effective pruning of task-trained recurrent neural networks using noisy fluctuations and connection rescalingcited by 0
  30. 2026Width-Independent Compressibility of Deep Neural Networkscited by 0
  31. 2026Theoretical Guarantees for One-Shot Magnitude Pruning and Compute-Adaptive Early Exitcited by 0