max-andr/relu_networks_overconfident
Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problem [CVPR 2019, oral]
Python
Stargazers
- ananth-jauntxr
- asusha94Poland/Gdańsk
- CodingBoOGuangdong University of Technology
- cwellszhangShanghai
- dxoigmnIntel Labs
- enneast
- Fancy-sfChina
- fly51flyPRIS
- gokriznasticNortheastern University
- huanglianghuaTongyi Lab
- huanzhang12University of Illinois
- iphysresearchICTP-AP
- j-cbTübingen, Germany
- kxhitImperial College, ZJU, HIT
- LiberiFataliVietnam
- lliai
- max-andrEPFL
- mayurbhangale@deliveryhero
- meetpsMountain View, CA
- mikigomKAIST
- msnqqerHuawei
- oleksandrlazariev
- pbamotraAutodesk, Ex-Coupang, CMU-Pittsburgh
- QL111111
- r06922019
- radao@scythe-robotics
- raphaelachddou
- real-Xuan
- SaintLogos1234
- ShangxuanWu
- suruoxiAlibaba
- TinkerMob杭州
- ToTheMonn
- wangxuanzhaoBUPT
- WeihongMSCUT
- zbessingerZillow Group