News

GitHub9d

A self-supervised clustering method combining adversarial autoencoders with cross-attention-based graph convolutional networks (GCN) for cell type identification in scRNA-seq ...

The code runs under Python 3.9.20 and Pytorch 2.3.1+cu121. Create a PyTorch environment and install required packages, such as "numpy", "pandas", "scikit-learn ...
IEEE16d

A Classifier Adaptation and Adversarial Learning Joint Framework for Cross-Scene Coastal Wetland Mapping on Hyperspectral Imagery

Abstract: Coastal wetlands play a crucial role in maintaining global ecological security and biodiversity. Hyperspectral image (HSI), with its rich spectral information and high spatial resolution, ...
IEEE25d

A Unified Framework for Generating Diverse and Stealthy Adversarial Patches Against Aerial Object Detection

Abstract: Deep neural network (DNN) has become important in the aerial object detection field. However, due to the vulnerabilities of DNN, adversarial patch is proved to be an efficient method to ...
GitHub26d

Discrepancy between PyTorch and TensorRT when casting from FP32 to BF16 causes significant accuracy mismatch

I encountered a reproducibility issue when exporting a PyTorch model to ONNX and running it with TensorRT in bfloat16 mode. Although the computation logic and operator mapping between PyTorch and ...
Hosted on MSN27d

Lawyer 'very confident' a foreign adversary attacked Canadian diplomats in Cuba

OTTAWA — A lawyer for Canadian diplomats and their families says he believes the mysterious ailments they suffered in Cuba were caused by a foreign adversary, despite a federal government report that ...