Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Rewrite the Stars

About

Recent studies have drawn attention to the untapped potential of the "star operation" (element-wise multiplication) in network design. While intuitive explanations abound, the foundational rationale behind its application remains largely unexplored. Our study attempts to reveal the star operation's ability to map inputs into high-dimensional, non-linear feature spaces -- akin to kernel tricks -- without widening the network. We further introduce StarNet, a simple yet powerful prototype, demonstrating impressive performance and low latency under compact network structure and efficient budget. Like stars in the sky, the star operation appears unremarkable but holds a vast universe of potential. Our work encourages further exploration across tasks, with codes available at https://github.com/ma-xu/Rewrite-the-Stars.

Xu Ma, Xiyang Dai, Yue Bai, Yizhou Wang, Yun Fu• 2024

Related benchmarks

TaskDatasetResultRank
Semantic segmentationADE20K (val)
mIoU36
2731
Object DetectionCOCO 2017
AP (Box)33.8
279
Instance SegmentationCOCO 2017
APm31.9
199
Image ClassificationImageNet-C (val)
mCE77.5
97
Image ClassificationImageNet-R (val)
Accuracy34.1
82
Image ClassificationImageNet-1K 1.0 (val)
Top-1 Acc78.4
55
Image ClassificationImageNet-A (val)--
55
Image ClassificationImageNet-Sketch (val)
Top-1 Acc21.8
52
Showing 8 of 8 rows

Other info

Code

Follow for update