mlsanityData to 7 Oct 2026

LOss-Based SensiTivity rEgulaRization: towards deep sparse neural networks

arXiv 2011.09905 ↗First posted 16 Nov 2020cs.LGcited by 19

Not on the map. Why?

Abstract (arXiv)

LOBSTER (LOss-Based SensiTivity rEgulaRization) is a method for training neural networks having a sparse topology. Let the sensitivity of a network parameter be the variation of the loss function with respect to the variation of the parameter. Parameters with low sensitivity, i.e. having little impact on the loss when perturbed, are shrunk and then pruned to sparsify the network. Our method allows to train a network from scratch, i.e. without preliminary learning or rewinding. Experiments on multiple architectures and datasets show competitive compression ratios with minimal computational overhead.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). First posted on arXiv 16 Nov 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 6; 2023: 4; 2024: 4; 2025: 1; 2026: 2.

19 papers.

Sort:

  1. 2022Packed-Ensembles for Efficient Uncertainty Estimationcited by 49
  2. 2022Compressing Explicit Voxel Grid Representations: fast NeRFs become also smallcited by 48
  3. 2023Can Unstructured Pruning Reduce the Depth in Deep Neural Networks?cited by 20
  4. 2023Adaptive Sharpness-Aware Pruning for Robust Sparse Networkscited by 17
  5. 2022To update or not to update? Neurons at equilibrium in deep modelscited by 13
  6. 2021Adaptive Precision Training (AdaPT): A dynamic fixed point quantized training approach for DNNscited by 8
  7. 2023DSD2: Can We Dodge Sparse Double Descent and Compress the Neural Network Worry-Free?cited by 6
  8. 2021HEMP: High-order Entropy Minimization for neural network comPressioncited by 4
  9. 2024Layer Collapse Can be Induced by Unstructured Pruningcited by 4
  10. 2022Regularization-based Pruning of Irrelevant Weights in Deep Neural Architecturescited by 3
  11. 2022Towards Efficient Capsule Networkscited by 3
  12. 2024The Simpler The Better: An Entropy-Based Importance Metric To Reduce Neural Networks' Depthcited by 3
7 more
  1. 2022Capsule Networks Do Not Need to Model Everythingcited by 2
  2. 2024SkinFormer: Learning Statistical Texture Representation with Transformer for Skin Lesion Segmentationcited by 2
  3. 2024LaCoOT: Layer Collapse through Optimal Transportcited by 1
  4. 2025Efficient Adaptation of Deep Neural Networks for Semantic Segmentation in Space Applicationscited by 1
  5. 2026Routing the Lottery: Adaptive Subnetworks for Heterogeneous Datacited by 1
  6. 2023Learn how to Prune Pixels for Multi-view Neural Image-based Synthesiscited by 0
  7. 2026From Deep to Shallow: Unconstrained and Efficient Layer Merging Strategycited by 0