Sparser, Better, Deeper, Stronger: Improving Sparse Training with Exact Orthogonal Initialization

• URL: http://arxiv.org/abs/2406.01755v1
• Date: Mon, 3 Jun 2024 19:44:47 GMT
• Title: Sparser, Better, Deeper, Stronger: Improving Sparse Training with Exact Orthogonal Initialization
• Authors: Aleksandra Irena Nowak, Łukasz Gniecki, Filip Szatkowski, Jacek Tabor,
• Abstract summary: Static sparse training aims to train sparse models from scratch, achieving remarkable results in recent years. We propose Exact Orthogonal Initialization (EOI), a novel sparse Orthogonal Initialization scheme based on random Givens rotations. Our method enables training highly sparse 1000-layer and CNN networks without residual connections or normalization techniques.
• Score: 49.06421851486415
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Static sparse training aims to train sparse models from scratch, achieving remarkable results in recent years. A key design choice is given by the sparse initialization, which determines the trainable sub-network through a binary mask. Existing methods mainly select such mask based on a predefined dense initialization. Such an approach may not efficiently leverage the mask's potential impact on the optimization. An alternative direction, inspired by research into dynamical isometry, is to introduce orthogonality in the sparse subnetwork, which helps in stabilizing the gradient signal. In this work, we propose Exact Orthogonal Initialization (EOI), a novel sparse orthogonal initialization scheme based on composing random Givens rotations. Contrary to other existing approaches, our method provides exact (not approximated) orthogonality and enables the creation of layers with arbitrary densities. We demonstrate the superior effectiveness and efficiency of EOI through experiments, consistently outperforming common sparse initialization techniques. Our method enables training highly sparse 1000-layer MLP and CNN networks without residual connections or normalization techniques, emphasizing the crucial role of weight initialization in static sparse training alongside sparse mask selection. The code is available at https://github.com/woocash2/sparser-better-deeper-stronger

Related papers

• What to Prune and What Not to Prune at Initialization [0.0]
  Post-training dropout based approaches achieve high sparsity. Initialization pruning is more efficacious when it comes to scaling computation cost of the network. The goal is to achieve higher sparsity while preserving performance.
  arXiv  Detail & Related papers  (2022-09-06T03:48:10Z)
• Training Your Sparse Neural Network Better with Any Mask [106.134361318518]
  Pruning large neural networks to create high-quality, independently trainable sparse masks is desirable. In this paper we demonstrate an alternative opportunity: one can customize the sparse training techniques to deviate from the default dense network training protocols. Our new sparse training recipe is generally applicable to improving training from scratch with various sparse masks.
  arXiv  Detail & Related papers  (2022-06-26T00:37:33Z)
• Efficient Few-Shot Object Detection via Knowledge Inheritance [62.36414544915032]
  Few-shot object detection (FSOD) aims at learning a generic detector that can adapt to unseen tasks with scarce training samples. We present an efficient pretrain-transfer framework (PTF) baseline with no computational increment. We also propose an adaptive length re-scaling (ALR) strategy to alleviate the vector length inconsistency between the predicted novel weights and the pretrained base weights.
  arXiv  Detail & Related papers  (2022-03-23T06:24:31Z)
• Boosting Fast Adversarial Training with Learnable Adversarial Initialization [79.90495058040537]
  Adrial training (AT) has been demonstrated to be effective in improving model robustness by leveraging adversarial examples for training. To boost training efficiency, fast gradient sign method (FGSM) is adopted in fast AT methods by calculating gradient only once.
  arXiv  Detail & Related papers  (2021-10-11T05:37:00Z)
• Data-driven Weight Initialization with Sylvester Solvers [72.11163104763071]
  We propose a data-driven scheme to initialize the parameters of a deep neural network. We show that our proposed method is especially effective in few-shot and fine-tuning settings.
  arXiv  Detail & Related papers  (2021-05-02T07:33:16Z)
• An Effective and Efficient Initialization Scheme for Training Multi-layer Feedforward Neural Networks [5.161531917413708]
  We propose a novel network initialization scheme based on the celebrated Stein's identity. A proposed SteinGLM method is shown through extensive numerical results to be much faster and more accurate than other popular methods commonly used for training neural networks.
  arXiv  Detail & Related papers  (2020-05-16T16:17:37Z)
• MSE-Optimal Neural Network Initialization via Layer Fusion [68.72356718879428]
  Deep neural networks achieve state-of-the-art performance for a range of classification and inference tasks. The use of gradient combined nonvolutionity renders learning susceptible to novel problems. We propose fusing neighboring layers of deeper networks that are trained with random variables.
  arXiv  Detail & Related papers  (2020-01-28T18:25:15Z)

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.