Yin Li

Beyond Grids: Learning Graph Representations for Visual Recognition

2018 · Neural Information Processing Systems

We propose learning graph representations from 2D feature maps for visual recognition. Our method draws inspiration from region based recognition, and learns to transform a 2D image into a graph structure. The vertices of the …

AI-assisted super-resolution cosmological simulations IV: An emulator for deterministic realizations

2024 · arXiv (Cornell University)

Super-resolution (SR) models in cosmological simulations use deep learning (DL) to rapidly enhance low-resolution (LR) runs with statistically correct fine details. These models preserves large-scale structures by conditioning on an LR version of the simulation. …