Penghui Qi

4 papers · 2021–2025 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🗺️ Taxonomy Completionist (11) 🏆 Keyword Champion (2)

Venues

ICML (2) ICLR (1) NeurIPS (1)

Top co-authors

Min Lin (3) Xinyi Wan (3) Guangxing Huang (2) Weimin Li (1) Zhenkun Tang (1) Quan Yuan (1) Xiongjun Pi (1) Junxiao Song (1) Haitao Long (1) Nyamdavaa Amar (1)

Research topics

Machine Learning (3) Application Areas (3) Optimization & Theory · Machine Learning (3) Artificial Intelligence (2) Core AI (2) Deep Learning (1) Reinforcement Learning (1) Applications (1) Methods (1) Optimization & Theory · Deep Learning (1)

Keywords

pipeline parallelism (3) distributed training (2) activation memory (2) throughput optimization (2) memory optimization (1) starcraft ii (1) large language model (1) real-time strategy (1) neural network (1) hybrid parallelism (1) bubble reduction (1) optimizer synchronization (1) backward computation (1) acceleration (1) throughput (1) memory offload (1) deep reinforcement learning (1) microbatches (1) imitation learning (1) model parallelism (1)

Papers

PipeOffload: Improving Scalability of Pipeline Parallelism with Memory Optimization ICML 2025 Pipeline Parallelism with Controllable Memory NeurIPS 2024 Zero Bubble (Almost) Pipeline Parallelism ICLR 2024 SCC: an efficient deep reinforcement learning agent mastering the game of StarCraft II ICML 2021