Deep learning is transforming most areas of science and technology, including electron
microscopy. This review paper offers a practical perspective aimed at developers with …

Optimization is a critical component in deep learning. We think optimization for neural
networks is an interesting topic for theoretical research due to various reasons. First, its …

In this paper, we study Multiscale Vision Transformers (MViTv2) as a unified architecture for
image and video classification, as well as object detection. We present an improved version …

Image segmentation is often ambiguous at the level of individual image patches and
requires contextual information to reach label consensus. In this paper we introduce …

Abstract We present Multiscale Vision Transformers (MViT) for video and image recognition,
by connecting the seminal idea of multiscale feature hierarchies with transformer models …

Extracting robust feature representation is one of the key challenges in object re-
identification (ReID). Although convolution neural network (CNN)-based methods have …

Recently, neural networks purely based on attention were shown to address image
understanding tasks such as image classification. These high-performing vision …

In a series of recent theoretical works, it was shown that strongly over-parameterized neural
networks trained with gradient-based methods could converge exponentially fast to zero …

Finetuning a pretrained backbone in the encoder part of an image transformer network has
been the traditional approach for the semantic segmentation task. However, such an …

Federated learning (FL) is a heavily promoted approach for training ML models on sensitive
data, eg, text typed by users on their smartphones. FL is expressly designed for training on …