Elimination-compensation pruning for fully-connected neural networks

• URL: http://arxiv.org/abs/2602.20467v1
• Date: Tue, 24 Feb 2026 01:56:12 GMT
• Title: Elimination-compensation pruning for fully-connected neural networks
• Authors: Enrico Ballini, Luca Muscarnera, Alessio Fumagalli, Anna Scotti, Francesco Regazzoni,
• Abstract summary: Pruning techniques are used to extract sparse representations of neural networks parameters.<n>We introduce a novel method in which the importance measure of each weight is computed considering the output behavior.<n>Our findings are discussed and the theoretical implications of our results are presented.
• Score: 2.785400823678854
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: The unmatched ability of Deep Neural Networks in capturing complex patterns in large and noisy datasets is often associated with their large hypothesis space, and consequently to the vast amount of parameters that characterize model architectures. Pruning techniques affirmed themselves as valid tools to extract sparse representations of neural networks parameters, carefully balancing between compression and preservation of information. However, a fundamental assumption behind pruning is that expendable weights should have small impact on the error of the network, while highly important weights should tend to have a larger influence on the inference. We argue that this idea could be generalized; what if a weight is not simply removed but also compensated with a perturbation of the adjacent bias, which does not contribute to the network sparsity? Our work introduces a novel pruning method in which the importance measure of each weight is computed considering the output behavior after an optimal perturbation of its adjacent bias, efficiently computable by automatic differentiation. These perturbations can be then applied directly after the removal of each weight, independently of each other. After deriving analytical expressions for the aforementioned quantities, numerical experiments are conducted to benchmark this technique against some of the most popular pruning strategies, demonstrating an intrinsic efficiency of the proposed approach in very diverse machine learning scenarios. Finally, our findings are discussed and the theoretical implications of our results are presented.

Related papers

• Network reconstruction via the minimum description length principle [0.0]
  We propose an alternative nonparametric regularization scheme based on hierarchical Bayesian inference and weight quantization.<n>Our approach follows the minimum description length (MDL) principle, and uncovers the weight distribution that allows for the most compression of the data.<n>We demonstrate that our scheme yields systematically increased accuracy in the reconstruction of both artificial and empirical networks.
  arXiv  Detail & Related papers  (2024-05-02T05:35:09Z)
• On the Dynamics Under the Unhinged Loss and Beyond [104.49565602940699]
  We introduce the unhinged loss, a concise loss function, that offers more mathematical opportunities to analyze closed-form dynamics. The unhinged loss allows for considering more practical techniques, such as time-vary learning rates and feature normalization.
  arXiv  Detail & Related papers  (2023-12-13T02:11:07Z)
• Implicit Compressibility of Overparametrized Neural Networks Trained with Heavy-Tailed SGD [31.61477313262589]
  We consider a one-hidden-layer neural network trained with gradient descent (SGD) We show that if we inject additive heavy-tailed noise to the iterates at each, for any compression rate, there exists a level of overparametrization such that the output of the algorithm will be compressible with high probability.
  arXiv  Detail & Related papers  (2023-06-13T20:37:02Z)
• Bias in Pruned Vision Models: In-Depth Analysis and Countermeasures [93.17009514112702]
  Pruning, setting a significant subset of the parameters of a neural network to zero, is one of the most popular methods of model compression. Despite existing evidence for this phenomenon, the relationship between neural network pruning and induced bias is not well-understood.
  arXiv  Detail & Related papers  (2023-04-25T07:42:06Z)
• Formalizing Generalization and Robustness of Neural Networks to Weight Perturbations [58.731070632586594]
  We provide the first formal analysis for feed-forward neural networks with non-negative monotone activation functions against weight perturbations. We also design a new theory-driven loss function for training generalizable and robust neural networks against weight perturbations.
  arXiv  Detail & Related papers  (2021-03-03T06:17:03Z)
• Non-Singular Adversarial Robustness of Neural Networks [58.731070632586594]
  Adrial robustness has become an emerging challenge for neural network owing to its over-sensitivity to small input perturbations. We formalize the notion of non-singular adversarial robustness for neural networks through the lens of joint perturbations to data inputs as well as model weights.
  arXiv  Detail & Related papers  (2021-02-23T20:59:30Z)
• Tensor-Train Networks for Learning Predictive Modeling of Multidimensional Data [0.0]
  A promising strategy is based on tensor networks, which have been very successful in physical and chemical applications. We show that the weights of a multidimensional regression model can be learned by means of tensor networks with the aim of performing a powerful compact representation. An algorithm based on alternating least squares has been proposed for approximating the weights in TT-format with a reduction of computational power.
  arXiv  Detail & Related papers  (2021-01-22T16:14:38Z)
• Stable Recovery of Entangled Weights: Towards Robust Identification of Deep Neural Networks from Minimal Samples [0.0]
  We introduce the so-called entangled weights, which compose weights of successive layers intertwined with suitable diagonal and invertible matrices depending on the activation functions and their shifts. We prove that entangled weights are completely and stably approximated by an efficient and robust algorithm. In terms of practical impact, our study shows that we can relate input-output information uniquely and stably to network parameters, providing a form of explainability.
  arXiv  Detail & Related papers  (2021-01-18T16:31:19Z)
• Loss Bounds for Approximate Influence-Based Abstraction [81.13024471616417]
  Influence-based abstraction aims to gain leverage by modeling local subproblems together with the 'influence' that the rest of the system exerts on them. This paper investigates the performance of such approaches from a theoretical perspective. We show that neural networks trained with cross entropy are well suited to learn approximate influence representations.
  arXiv  Detail & Related papers  (2020-11-03T15:33:10Z)
• A Partial Regularization Method for Network Compression [0.0]
  We propose an approach of partial regularization rather than the original form of penalizing all parameters, which is said to be full regularization, to conduct model compression at a higher speed. Experimental results show that as we expected, the computational complexity is reduced by observing less running time in almost all situations. Surprisingly, it helps to improve some important metrics such as regression fitting results and classification accuracy in both training and test phases on multiple datasets.
  arXiv  Detail & Related papers  (2020-09-03T00:38:27Z)
• Understanding Generalization in Deep Learning via Tensor Methods [53.808840694241]
  We advance the understanding of the relations between the network's architecture and its generalizability from the compression perspective. We propose a series of intuitive, data-dependent and easily-measurable properties that tightly characterize the compressibility and generalizability of neural networks.
  arXiv  Detail & Related papers  (2020-01-14T22:26:57Z)

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.