L-SWAG: Layer-Sample Wise Activation with Gradients information for Zero-Shot NAS on Vision Transformers

• URL: http://arxiv.org/abs/2505.07300v1
• Date: Mon, 12 May 2025 07:44:52 GMT
• Title: L-SWAG: Layer-Sample Wise Activation with Gradients information for Zero-Shot NAS on Vision Transformers
• Authors: Sofia Casarin, Sergio Escalera, Oswald Lanz,
• Abstract summary: Training-free Neural Architecture Search (NAS) efficiently identifies high-performing neural networks using zero-cost (ZC) proxies.<n>ZC-NAS is both (i) time-efficient, eliminating the need for model training, and (ii) interpretable, with proxy designs often theoretically grounded.<n>This work extends ZC proxy applicability to Vision Transformers (ViTs)
• Score: 39.19675815138566
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Training-free Neural Architecture Search (NAS) efficiently identifies high-performing neural networks using zero-cost (ZC) proxies. Unlike multi-shot and one-shot NAS approaches, ZC-NAS is both (i) time-efficient, eliminating the need for model training, and (ii) interpretable, with proxy designs often theoretically grounded. Despite rapid developments in the field, current SOTA ZC proxies are typically constrained to well-established convolutional search spaces. With the rise of Large Language Models shaping the future of deep learning, this work extends ZC proxy applicability to Vision Transformers (ViTs). We present a new benchmark using the Autoformer search space evaluated on 6 distinct tasks and propose Layer-Sample Wise Activation with Gradients information (L-SWAG), a novel, generalizable metric that characterizes both convolutional and transformer architectures across 14 tasks. Additionally, previous works highlighted how different proxies contain complementary information, motivating the need for a ML model to identify useful combinations. To further enhance ZC-NAS, we therefore introduce LIBRA-NAS (Low Information gain and Bias Re-Alignment), a method that strategically combines proxies to best represent a specific benchmark. Integrated into the NAS search, LIBRA-NAS outperforms evolution and gradient-based NAS techniques by identifying an architecture with a 17.0% test error on ImageNet1k in just 0.1 GPU days.

Related papers

• TG-NAS: Leveraging Zero-Cost Proxies with Transformer and Graph Convolution Networks for Efficient Neural Architecture Search [1.30891455653235]
  TG-NAS aims to create training-free proxies for architecture performance prediction. We introduce TG-NAS, a novel model-based universal proxy that leverages a transformer-based operator embedding generator and a graph convolution network (GCN) to predict architecture performance. TG-NAS achieves up to 300X improvements in search efficiency compared to previous SOTA ZC proxy methods.
  arXiv  Detail & Related papers  (2024-03-30T07:25:30Z)
• DNA Family: Boosting Weight-Sharing NAS with Block-Wise Supervisions [121.05720140641189]
  We develop a family of models with the distilling neural architecture (DNA) techniques. Our proposed DNA models can rate all architecture candidates, as opposed to previous works that can only access a sub- search space using algorithms. Our models achieve state-of-the-art top-1 accuracy of 78.9% and 83.6% on ImageNet for a mobile convolutional network and a small vision transformer, respectively.
  arXiv  Detail & Related papers  (2024-03-02T22:16:47Z)
• SiGeo: Sub-One-Shot NAS via Information Theory and Geometry of Loss Landscape [14.550053893504764]
  We introduce a "sub-one-shot" paradigm that serves as a bridge between zero-shot and one-shot NAS. In sub-one-shot NAS, the supernet is trained using only a small subset of the training data, a phase we refer to as "warm-up" We present SiGeo, a proxy founded on a novel theoretical framework that connects the supernet warm-up with the efficacy of the proxy.
  arXiv  Detail & Related papers  (2023-11-22T05:25:24Z)
• DiffusionNAG: Predictor-guided Neural Architecture Generation with Diffusion Models [56.584561770857306]
  We propose a novel conditional Neural Architecture Generation (NAG) framework based on diffusion models, dubbed DiffusionNAG. Specifically, we consider the neural architectures as directed graphs and propose a graph diffusion model for generating them. We validate the effectiveness of DiffusionNAG through extensive experiments in two predictor-based NAS scenarios: Transferable NAS and Bayesian Optimization (BO)-based NAS. When integrated into a BO-based algorithm, DiffusionNAG outperforms existing BO-based NAS approaches, particularly in the large MobileNetV3 search space on the ImageNet 1K dataset.
  arXiv  Detail & Related papers  (2023-05-26T13:58:18Z)
• NASiam: Efficient Representation Learning using Neural Architecture Search for Siamese Networks [76.8112416450677]
  Siamese networks are one of the most trending methods to achieve self-supervised visual representation learning (SSL) NASiam is a novel approach that uses for the first time differentiable NAS to improve the multilayer perceptron projector and predictor (encoder/predictor pair) NASiam reaches competitive performance in both small-scale (i.e., CIFAR-10/CIFAR-100) and large-scale (i.e., ImageNet) image classification datasets while costing only a few GPU hours.
  arXiv  Detail & Related papers  (2023-01-31T19:48:37Z)
• Lightweight Neural Architecture Search for Temporal Convolutional Networks at the Edge [21.72253397805102]
  This work focuses in particular on Temporal Convolutional Networks (TCNs), a convolutional model for time-series processing. We propose the first NAS tool that explicitly targets the optimization of the most peculiar architectural parameters of TCNs. We test the proposed NAS on four real-world, edge-relevant tasks, involving audio and bio-signals.
  arXiv  Detail & Related papers  (2023-01-24T19:47:40Z)
• Towards Self-supervised and Weight-preserving Neural Architecture Search [38.497608743382145]
  We propose the self-supervised and weight-preserving neural architecture search (SSWP-NAS) as an extension of the current NAS framework. Experiments show that the architectures searched by the proposed framework achieve state-of-the-art accuracy on CIFAR-10, CIFAR-100, and ImageNet datasets.
  arXiv  Detail & Related papers  (2022-06-08T18:48:05Z)
• Generic Neural Architecture Search via Regression [27.78105839644199]
  We propose a novel and generic neural architecture search (NAS) framework, termed Generic NAS (GenNAS) GenNAS does not use task-specific labels but instead adopts textitregression on a set of manually designed synthetic signal bases for architecture evaluation. We then propose an automatic task search to optimize the combination of synthetic signals using limited downstream-task-specific labels.
  arXiv  Detail & Related papers  (2021-08-04T08:21:12Z)
• Binarized Neural Architecture Search for Efficient Object Recognition [120.23378346337311]
  Binarized neural architecture search (BNAS) produces extremely compressed models to reduce huge computational cost on embedded devices for edge computing. An accuracy of $96.53%$ vs. $97.22%$ is achieved on the CIFAR-10 dataset, but with a significantly compressed model, and a $40%$ faster search than the state-of-the-art PC-DARTS.
  arXiv  Detail & Related papers  (2020-09-08T15:51:23Z)
• DDPNAS: Efficient Neural Architecture Search via Dynamic Distribution Pruning [135.27931587381596]
  We propose an efficient and unified NAS framework termed DDPNAS via dynamic distribution pruning. In particular, we first sample architectures from a joint categorical distribution. Then the search space is dynamically pruned and its distribution is updated every few epochs. With the proposed efficient network generation method, we directly obtain the optimal neural architectures on given constraints.
  arXiv  Detail & Related papers  (2019-05-28T06:35:52Z)

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.