Mesoscale eddies with horizontal scales from tens to hundreds of kilometers are ubiquitous in the upper ocean, dominating the ocean variability from daily to weekly time scales. Their turbulent nature causes great scientific challenges and computational burdens in accurately forecasting the short-term evolution of the ocean states based on conventional physics-driven numerical models. Recently, artificial intelligence (AI)-based methods have achieved competitive forecast performance and greatly increased computational efficiency in weather forecasts, compared to numerical models. Yet, their application to ocean forecasts remains challenging due to the different dynamic characteristics of the atmosphere and the ocean. Here, we develop WenHai, a data-driven eddy-resolving global ocean forecast system (GOFS), by training a deep neural network (DNN). The bulk formulae on momentum, heat, and freshwater fluxes are incorporated into the DNN to improve the representation of air-sea interactions. Ocean dynamics is exploited in the DNN architecture design to preserve ocean mesoscale eddy variability. WenHai outperforms a state-of-the-art eddy-resolving numerical GOFS and AI-based GOFS for the temperature profile, salinity profile, sea surface temperature, sea level anomaly, and near-surface current forecasts led by 1 day to at least 10 days. Our results highlight expertise-guided deep learning as a promising pathway for enhancing the global ocean forecast capacity.
The success of image classification depends on copious annotated images for training. Annotating histopathology images is costly and laborious. Although several successful self-supervised representation learning approaches have been introduced, they are still insufficient to consider the unique characteristics of histopathology images. In this work, we propose the novel histopathology-oriented self-supervised representation learning framework (HistoSSL) to efficiently extract representations from unlabeled histopathology images at three levels: global, cell, and stain. The model transfers remarkably to downstream tasks: colorectal tissue phenotyping on the NCTCRC dataset and breast cancer metastasis recognition on the CAMELYON16 dataset. HistoSSL achieved higher accuracies than state-of-the-art self-supervised learning approaches, which proved the robustness of the learned representations.
25 January 2022 in IEEE Transactions on Parallel and Distributed Systems
Efficient numerical methods are promising tools for delivering unique insights into the fascinating properties of physics, such as the highly frustrated quantum many-body systems. However, the computational complexity of obtaining the wave functions for accurately describing the quantum states increases exponentially with respect to particle number. Here we present a novel convolutional neural network (CNN) for simulating the two-dimensional highly frustrated spin- $1/2$$J_1-J_2$ Heisenberg model, meanwhile the simulation is performed at an extreme scale system with low cost and high scalability. By ingenious employment of transfer learning and CNN’s translational invariance, we successfully investigate the quantum system with the lattice size up to $24\times 24$ , within 30 million cores of the new generation of sunway supercomputer. The final achievement demonstrates the effectiveness of CNN-based representation of quantum-state and brings the state-of-the-art record up to a brand-new level from both aspects of remarkable accuracy and unprecedented scales.
19 June 2018 in IEEE Transactions on Parallel and Distributed Systems
The study of strongly frustrated magnetic systems has drawn great attentions from both theoretical and experimental physics. Efficient simulations of these models are essential for understanding their exotic properties. Here we present PEPS++, a novel computational paradigm for simulating frustrated magnetic systems and other strongly correlated quantum many-body systems. PEPS++ can accurately solve these models at the extreme scale with low cost and high scalability on modern heterogeneous supercomputers. We implement PEPS++ on Sunway TaihuLight based on a carefully designed tensor computation library for manipulating high-rank tensors and optimize it by invoking various high-performance matrix and tensor operations. By solving a 2D strongly frustrated $J_1-J_2$ model with over ten million cores, PEPS++ demonstrates the capability of simulating strongly correlated quantum many-body problems at unprecedented scales with accuracy and time-to-solution far beyond the previous state of the art.