Reconstruction Task Finds Universal Winning Tickets
- URL: http://arxiv.org/abs/2202.11484v1
- Date: Wed, 23 Feb 2022 13:04:32 GMT
- Title: Reconstruction Task Finds Universal Winning Tickets
- Authors: Ruichen Li, Binghui Li, Qi Qian, Liwei Wang
- Abstract summary: Pruning well-trained neural networks is effective to achieve a promising accuracy-efficiency trade-off in computer vision regimes.
Most of existing pruning algorithms only focus on the classification task defined on the source domain.
In this paper, we show that the image-level pretrain task is not capable of pruning models for diverse downstream tasks.
- Score: 24.52604301906691
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Pruning well-trained neural networks is effective to achieve a promising
accuracy-efficiency trade-off in computer vision regimes. However, most of
existing pruning algorithms only focus on the classification task defined on
the source domain. Different from the strong transferability of the original
model, a pruned network is hard to transfer to complicated downstream tasks
such as object detection arXiv:arch-ive/2012.04643. In this paper, we show that
the image-level pretrain task is not capable of pruning models for diverse
downstream tasks. To mitigate this problem, we introduce image reconstruction,
a pixel-level task, into the traditional pruning framework. Concretely, an
autoencoder is trained based on the original model, and then the pruning
process is optimized with both autoencoder and classification losses. The
empirical study on benchmark downstream tasks shows that the proposed method
can outperform state-of-the-art results explicitly.
Related papers
- One-Shot Pruning for Fast-adapting Pre-trained Models on Devices [28.696989086706186]
Large-scale pre-trained models have been remarkably successful in resolving downstream tasks.
deploying these models on low-capability devices still requires an effective approach, such as model pruning.
We present a scalable one-shot pruning method that leverages pruned knowledge of similar tasks to extract a sub-network from the pre-trained model for a new task.
arXiv Detail & Related papers (2023-07-10T06:44:47Z) - Task-Robust Pre-Training for Worst-Case Downstream Adaptation [62.05108162160981]
Pre-training has achieved remarkable success when transferred to downstream tasks.
This paper considers pre-training a model that guarantees a uniformly good performance over the downstream tasks.
arXiv Detail & Related papers (2023-06-21T07:43:23Z) - GRIG: Few-Shot Generative Residual Image Inpainting [27.252855062283825]
We present a novel few-shot generative residual image inpainting method that produces high-quality inpainting results.
The core idea is to propose an iterative residual reasoning method that incorporates Convolutional Neural Networks (CNNs) for feature extraction.
We also propose a novel forgery-patch adversarial training strategy to create faithful textures and detailed appearances.
arXiv Detail & Related papers (2023-04-24T12:19:06Z) - Iterative Soft Shrinkage Learning for Efficient Image Super-Resolution [91.3781512926942]
Image super-resolution (SR) has witnessed extensive neural network designs from CNN to transformer architectures.
This work investigates the potential of network pruning for super-resolution iteration to take advantage of off-the-shelf network designs and reduce the underlying computational overhead.
We propose a novel Iterative Soft Shrinkage-Percentage (ISS-P) method by optimizing the sparse structure of a randomly network at each and tweaking unimportant weights with a small amount proportional to the magnitude scale on-the-fly.
arXiv Detail & Related papers (2023-03-16T21:06:13Z) - Attentive Fine-Grained Structured Sparsity for Image Restoration [63.35887911506264]
N:M structured pruning has appeared as one of the effective and practical pruning approaches for making the model efficient with the accuracy constraint.
We propose a novel pruning method that determines the pruning ratio for N:M structured sparsity at each layer.
arXiv Detail & Related papers (2022-04-26T12:44:55Z) - Joint Learning of Neural Transfer and Architecture Adaptation for Image
Recognition [77.95361323613147]
Current state-of-the-art visual recognition systems rely on pretraining a neural network on a large-scale dataset and finetuning the network weights on a smaller dataset.
In this work, we prove that dynamically adapting network architectures tailored for each domain task along with weight finetuning benefits in both efficiency and effectiveness.
Our method can be easily generalized to an unsupervised paradigm by replacing supernet training with self-supervised learning in the source domain tasks and performing linear evaluation in the downstream tasks.
arXiv Detail & Related papers (2021-03-31T08:15:17Z) - Cross-modal Adversarial Reprogramming [12.467311480726702]
Recent works on adversarial reprogramming have shown that it is possible to repurpose neural networks for alternate tasks without modifying the network architecture or parameters.
We analyze the feasibility of adversarially repurposing image classification neural networks for Natural Language Processing (NLP) and other sequence classification tasks.
arXiv Detail & Related papers (2021-02-15T03:46:16Z) - Counterfactual Generative Networks [59.080843365828756]
We propose to decompose the image generation process into independent causal mechanisms that we train without direct supervision.
By exploiting appropriate inductive biases, these mechanisms disentangle object shape, object texture, and background.
We show that the counterfactual images can improve out-of-distribution with a marginal drop in performance on the original classification task.
arXiv Detail & Related papers (2021-01-15T10:23:12Z) - Mixed-Privacy Forgetting in Deep Networks [114.3840147070712]
We show that the influence of a subset of the training samples can be removed from the weights of a network trained on large-scale image classification tasks.
Inspired by real-world applications of forgetting techniques, we introduce a novel notion of forgetting in mixed-privacy setting.
We show that our method allows forgetting without having to trade off the model accuracy.
arXiv Detail & Related papers (2020-12-24T19:34:56Z) - Principled network extraction from images [0.0]
We present a principled model to extract network topologies from images that is scalable and efficient.
We test our model on real images of the retinal vascular system, slime mold and river networks.
arXiv Detail & Related papers (2020-12-23T15:56:09Z) - Multi-task pre-training of deep neural networks for digital pathology [8.74883469030132]
We first assemble and transform many digital pathology datasets into a pool of 22 classification tasks and almost 900k images.
We show that our models used as feature extractors either improve significantly over ImageNet pre-trained models or provide comparable performance.
arXiv Detail & Related papers (2020-05-05T08:50:17Z)
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.