2026
Non publié
- Beyond All-Reduce: Event-Driven Model Parallelism Without Collective Communication Primitives (EBD2N). , working paper or preprint. doi www
2023
Articles de conférence
- Fully Distributed Deep Neural Network : F2D2N. In Mobile, Secure, and Programmable Networking. MSPN 2023, Springer, Paris, France, 2023. doi www