I
Ilija Radosavovic
Researcher at Facebook
Publications - 18
Citations - 2265
Ilija Radosavovic is an academic researcher from Facebook. The author has contributed to research in topics: Computer science & Network planning and design. The author has an hindex of 8, co-authored 12 publications receiving 939 citations. Previous affiliations of Ilija Radosavovic include University of California, Berkeley.
Papers
More filters
Proceedings ArticleDOI
Designing Network Design Spaces
TL;DR: The RegNet design space provides simple and fast networks that work well across a wide range of flop regimes, and outperform the popular EfficientNet models while being up to 5x faster on GPUs.
Proceedings ArticleDOI
Data Distillation: Towards Omni-Supervised Learning
TL;DR: It is argued that visual recognition models have recently become accurate enough that it is now possible to apply classic ideas about self-training to challenging real-world data and propose data distillation, a method that ensembles predictions from multiple transformations of unlabeled data, using a single model, to automatically generate new training annotations.
Posted Content
Data Distillation: Towards Omni-Supervised Learning
TL;DR: Data distillation as mentioned in this paper ensembles predictions from multiple transformations of unlabeled data, using a single model, to automatically generate new training annotations, which can be used for omni-supervised learning.
Proceedings ArticleDOI
Attentive Single-Tasking of Multiple Tasks
TL;DR: In this paper, the authors propose a single-tasking multiple tasks (SMMT) model, where a network is trained on multiple tasks, but performs one task at a time.
Posted Content
Designing Network Design Spaces
TL;DR: In this paper, the authors propose a new network design paradigm called RegNet, where instead of focusing on designing individual network instances, they design network design spaces that parametrize populations of networks.