Matthias Reso
Employers on the program
- 2025Meta1 talk
Events1
Conferences1
Talks1
Years active2025
Appearances by year
- PyTorch Conference
Talks
| Date | Event | Conference | Role | Title | Links |
|---|---|---|---|---|---|
| Oct 22, 2025 | PyTorch Conference 2025 | PyTorch Conference | speaker | Efficient MoE Pre-training at Scale on AMD GPUs With TorchTitan as Meta Mixture-of-Experts (MoE) architectures enable efficient scaling of deep learning models but face challenges across diverse hardware. We present a scalable,…Show lessMixture-of-Experts (MoE) architectures enable efficient scaling of deep learning models but face challenges across diverse hardware. We present a scalable, optimized MoE pre-training pipeline with TorchTitan on AMD MI300X GPUs, built fully within PyTorch ecosystem. Our method uses efficient FP8... |
Frequent co-speakers
| Speaker | Shared events | Shared talks | Last together |
|---|---|---|---|
| Liz Li | 1 | 1 | 2025 |
| Yanyuan Qin | 1 | 1 | 2025 |
Speakers with the same name are merged into one entry, so this page may combine more than one person. Affiliations are shown as each conference listed them.