64 citations · 284 across the 15 of their papers we have counts for
16 papers
Solving ImageNet: a Unified Scheme for Training any Backbone to Top Results
Tal Ridnik, Hussam Lawen, Emanuel Ben-Baruch +1
ImageNet serves as the primary dataset for evaluating the quality of computer-vision models. The common practice today is training each architecture with a tailor-made scheme, desi…
Diverse Imagenet Models Transfer Better
Niv Nayman, Avram Golbert, Asaf Noy +2
A commonly accepted hypothesis is that models with higher accuracy on Imagenet perform better on other downstream tasks, leading to much research dedicated to optimizing Imagenet a…
ML-Decoder: Scalable and Versatile Classification Head
Tal Ridnik, Gilad Sharir, Avi Ben-Cohen +2
In this paper, we introduce ML-Decoder, a new attention-based classification head. ML-Decoder predicts the existence of class labels via queries, and enables better utilization of…
Multi-label Classification with Partial Annotations using Class-aware Selective Loss
Emanuel Ben-Baruch, Tal Ridnik, Itamar Friedman +4
Large-scale multi-label classification datasets are commonly, and perhaps inevitably, partially annotated. That is, only a small subset of labels are annotated per sample. Differen…
BINAS: Bilinear Interpretable Neural Architecture Search
Niv Nayman, Yonathan Aflalo, Asaf Noy +2
Practical use of neural networks often involves requirements on latency, energy and memory among others. A popular approach to find networks under such requirements is through cons…
PETA: Photo Albums Event Recognition using Transformers Attention
Tamar Glaser, Emanuel Ben-Baruch, Gilad Sharir +3
In recent years the amounts of personal photos captured increased significantly, giving rise to new challenges in multi-image understanding and high-level image understanding. Even…
ImageNet-21K Pretraining for the Masses
Tal Ridnik, Emanuel Ben-Baruch, Asaf Noy +1
ImageNet-1K serves as the primary dataset for pretraining deep learning models for computer vision tasks. ImageNet-21K dataset, which is bigger and more diverse, is used less frequ…
An Image is Worth 16x16 Words, What is a Video Worth?
Gilad Sharir, Asaf Noy, Lihi Zelnik-Manor
Leading methods in the domain of action recognition try to distill information from both the spatial and temporal dimensions of an input video. Methods that reach State of the Art…
HardCoRe-NAS: Hard Constrained diffeRentiable Neural Architecture Search
Niv Nayman, Yonathan Aflalo, Asaf Noy +1
Realistic use of neural networks often requires adhering to multiple constraints on latency, energy and memory among others. A popular approach to find fitting networks is through…
A Convergence Theory Towards Practical Over-parameterized Deep Neural Networks
Asaf Noy, Yi Xu, Yonathan Aflalo +2
Deep neural networks' remarkable ability to correctly fit training data when optimized by gradient-based algorithms is yet to be fully understood. Recent theoretical results explai…
WeMix: How to Better Utilize Data Augmentation
Yi Xu, Asaf Noy, Ming Lin +3
Data augmentation is a widely used training trick in deep learning to improve the network generalization ability. Despite many encouraging results, several recent studies did point…
Asymmetric Loss For Multi-Label Classification
Emanuel Ben-Baruch, Tal Ridnik, Nadav Zamir +4
In a typical multi-label setting, a picture contains on average few positive labels, and many negative ones. This positive-negative imbalance dominates the optimization process, an…
TResNet: High Performance GPU-Dedicated Architecture
Tal Ridnik, Hussam Lawen, Asaf Noy +3
Many deep learning models, developed in recent years, reach higher ImageNet accuracy than ResNet50, with fewer or comparable FLOPS count. While FLOPs are often seen as a proxy for…
Knapsack Pruning with Inner Distillation
Yonathan Aflalo, Asaf Noy, Ming Lin +2
Neural network pruning reduces the computational cost of an over-parameterized network to improve its efficiency. Popular methods vary from -norm sparsification to Neural A…
XNAS: Neural Architecture Search with Expert Advice
Niv Nayman, Asaf Noy, Tal Ridnik +3
This paper introduces a novel optimization method for differential neural architecture search, based on the theory of prediction with expert advice. Its optimization criterion is w…
ASAP: Architecture Search, Anneal and Prune
Asaf Noy, Niv Nayman, Tal Ridnik +5
Automatic methods for Neural Architecture Search (NAS) have been shown to produce state-of-the-art network models. Yet, their main drawback is the computational complexity of the s…