Deep Model Transferability from Attribution Maps

September 26, 2019 ยท Entered Twilight ยท ๐Ÿ› Neural Information Processing Systems

๐ŸŒ… TWILIGHT: Old Age
Predates the code-sharing era โ€” a pioneer of its time

"Last commit was 6.0 years ago (โ‰ฅ5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .DS_Store, .gitattributes, LICENSE, README.md, dataset, demo, experiments, explain_result, lib, requirement.txt, tools, vipa-logo.png

Authors Jie Song, Yixin Chen, Xinchao Wang, Chengchao Shen, Mingli Song arXiv ID 1909.11902 Category cs.CV: Computer Vision Citations 60 Venue Neural Information Processing Systems Repository https://github.com/zju-vipa/TransferbilityFromAttributionMaps โญ 20 Last Checked 2 months ago
Abstract
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.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Computer Vision