Exploring the Limits of Out-of-Distribution Detection
Dec 6, 2021
Speakers
Stanislav Fort
Speaker · 3 followers
Balaji Lakshminarayanan
Speaker · 5 followers
About
Near out-of-distribution detection (OOD) is a major challenge for deep neural networks. We demonstrate that large-scale pre-trained transformers can significantly improve the state-of-the-art (SOTA) on a range of near OOD tasks across different data modalities. For instance, on CIFAR-100 vs CIFAR-10 OOD detection, we improve the AUROC from 85Near out-of-distribution detection (OOD) is a major challenge for deep neural networks. We demonstrate that large-scale pre-trained transformers can significantly improve the state-of-the-art (SOTA) on a range of near OOD tasks across different data modalities. For instance, on CIFAR-100 vs CIFAR-10 OOD detection, we improve the AUROC from 85…
Organizer
NeurIPS 2021
Account · 1.9k followers
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Balancing Structure In NeuroSymbolic Methods
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Prior-independent Dynamic Auctions for a Value-maximizing Buyer
Watch later
Yuan Deng, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Algorithms for data-centric AI
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Defuse: Training More Robust Models through Creation and Correction of Novel Model Errors
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Learning to identify cracks on wind turbine blade surfaces using drone-based inspection images
Watch later
Akshay Iyer, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%