Session Chair: Martin Maas (Google DeepMind)

Forecasting GPU Performance for Deep Learning Training and Inference
Seonho Lee (Georgia Institute of Technology), Amar Phanishayee (Meta), Divya Mahajan (Georgia Institute of Technology)
Paper

MVQ: Towards Efficient DNN Compression and Acceleration with Masked Vector Quantization
Shuaiting Li (Zhejiang University), Chengxuan Wang (Zhejiang University), Juncan Deng (Zhejiang University), Zeyu Wang (Zhejiang University), Zewen Ye (Zhejiang University), Zongsheng Wang (Zhejiang University), Haibin Shen (Zhejiang University), Kejie Huang (Zhejiang University)
Paper

PartIR: Composing SPMD Partitioning Strategies for Machine Learning
Sami Alabed (Google DeepMind), Daniel Belov (Google DeepMind), Bart Chrzaszcz (Google DeepMind), Juliana Franco (Google DeepMind), Dominik Grewe (Google DeepMind), Dougal Maclaurin (Google DeepMind), James Molloy (Google DeepMind), Tom Natan (Google DeepMind), Tamara Norman (Google DeepMind), Xiaoyue Pan (Google DeepMind), Adam Paszke (Google DeepMind), Norman Alexander Rink (Google DeepMind), Michael Schaarschmidt (Isomorphic Labs), Timur Sitdikov (Google DeepMind), Agnieszka Swietlik (Google DeepMind), Dimitrios Vytiniotis (Google DeepMind), Joel Wee (Google DeepMind)
Paper

Using Analytical Performance/Power Model and Fine-Grained DVFS to Enhance AI Accelerator Energy Efficiency
Zibo Wang (State Key Laboratory for Novel Software Technology, Nanjing University), Yijia Zhang (Peng Cheng Laboratory), Fuchun Wei (Huawei Technologies Co., Ltd), Bingqiang Wang (Peng Cheng Laboratory), Yanlin Liu (Huawei Technologies Co., Ltd), Zhiheng Hu (Huawei Technologies Co., Ltd), Jingyi Zhang (Huawei Technologies Co., Ltd), Xiaoxin Xu (Huawei Technologies Co., Ltd), Jian He (Huawei Technologies Co., Ltd), Xiaoliang Wang (State Key Laboratory for Novel Software Technology, Nanjing University), Wanchun Dou (State Key Laboratory for Novel Software Technology, Nanjing University), Guihai Chen (State Key Laboratory for Novel Software Technology, Nanjing University), Chen Tian (State Key Laboratory for Novel Software Technology, Nanjing University)
Paper

Early Termination for Hyperdimensional Computing Using Inferential Statistics
Pu (Luke) Yi (Stanford University), Yifan Yang (Stanford University), Chae Young Lee (Stanford University), Sara Achour (Stanford University)
Paper