Pytorch Model
PyTorch is a widely used open-source machine learning framework primarily focused on facilitating the development and deployment of deep learning models. Current research emphasizes improving efficiency and accessibility through optimized implementations of various algorithms (e.g., symbolic regression, bundle adjustment, spiking neural networks) and the creation of specialized toolkits for tasks like species distribution modeling, adversarial machine learning, and multi-objective optimization. This framework's impact stems from its ease of use, extensibility, and support for GPU acceleration, enabling researchers and practitioners across diverse scientific disciplines to leverage deep learning for complex problems.
Papers
PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel
Yanli Zhao, Andrew Gu, Rohan Varma, Liang Luo, Chien-Chin Huang, Min Xu, Less Wright, Hamid Shojanazeri, Myle Ott, Sam Shleifer, Alban Desmaison, Can Balioglu, Pritam Damania, Bernard Nguyen, Geeta Chauhan, Yuchen Hao, Ajit Mathews, Shen Li
SequeL: A Continual Learning Library in PyTorch and JAX
Nikolaos Dimitriadis, Francois Fleuret, Pascal Frossard