paper-with-me

Papers

Deep Model Transferability from Attribution Maps

2019-09-26 · NeurIPS 2019 12 · Jie Song, Yixin Chen, Xinchao Wang, Chengchao Shen, Mingli Song

Exploring the transferability between heterogeneous tasks sheds light on their intrinsic interconnections, and consequently enables knowledge transfer from one task to another so as to reduce the training effort of the latter. In this paper, we propose an embarrassingly simple yet very efficacious approach to estimating the transferability of deep networks, especially those handling vision tasks. Unlike the seminal work of taskonomy that relies on a large number of annotations as supervision and is thus computationally cumbersome, the proposed approach requires no human annotations and imposes no constraints on the architectures of the networks. This is achieved, specifically, via projecting deep networks into a model space, wherein each network is treated as a point and the distances between two points are measured by deviations of their produced attribution maps. The proposed approach is several-magnitude times faster than taskonomy, and meanwhile preserves a task-wise topological structure highly similar to the one obtained by taskonomy. Code is available at https://github.com/zju-vipa/TransferbilityFromAttributionMaps.

📄 PDF Abstract BibTeX arXiv:1909.11902

Code (2)

zju-vipa/TransferbilityFromAttributionMaps 공식 구현 tf
zju-vipa/KamalEngine pytorch

Tasks

modelTransfer Learning

Similar Papers 제목 키워드 기반

DEPARA: Deep Attribution Graph for Deep Knowledge Transferability

2020-03-17 · CVPR 2020 6 · Jie Song, Yixin Chen, Jingwen Ye, Xinchao Wang 외

Exploring the intrinsic interconnections between the knowledge encoded in PRe-trained Deep Neural Networks (PR-DNNs) of heterogeneous tasks sheds light on their mutual transferability, and consequently enables knowledge …

Model SelectionTransfer Learning

Avengers Ensemble! Improving Transferability of Authorship Obfuscation

2021-09-15 · Muhammad Haroon, Fareed Zaffar, Padmini Srinivasan, Zubair Shafiq

Stylometric approaches have been shown to be quite effective for real-world authorship attribution. To mitigate the privacy threat posed by authorship attribution, researchers have proposed automated authorship obfuscati…

Authorship Attribution

Frequency-Aware Model Parameter Explorer: A new attribution method for improving explainability

2025-09-25 · Ali Yavari, Alireza Mohamadi, Elham Beydaghi, Philipp Seeböck 외 arxiv

State-of-the-art attribution methods rely on adversarial sample generation that applies an all-pass filter across the frequency spectrum, discarding fine-grained high-frequency information that is demonstrably important …

Improving Adversarial Transferability via Neuron Attribution-Based Attacks

2022-03-31 · CVPR 2022 1 · Jianping Zhang, Weibin Wu, Jen-tse Huang, Yizhan Huang 외

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples. It is thus imperative to devise effective attack algorithms to identify the deficiencies of DNNs beforehand in security-sensitive applicatio…

Attribute

Evaluating Feature Attribution Methods in the Image Domain

2022-02-22 · Arne Gevaert, Axel-Jan Rousseau, Thijs Becker, Dirk Valkenborg 외

Feature attribution maps are a popular approach to highlight the most important pixels in an image for a given prediction of a model. Despite a recent growth in popularity and available methods, little attention is given…

Benchmarking