Yann N. Dauphin
Research Scientist · Google DeepMind
About
Yann N. Dauphin is a machine learning researcher at Google DeepMind working on advancing deep learning algorithms. His work has introduced several seminal techniques in modern AI, including Gated Linear Units (GLU), the Mixup data augmentation method, and Top-K sampling for large language models. Before joining Google, he was a Research Scientist at Facebook AI Research. He completed his PhD in 2016 at the University of Montreal under the supervision of Turing Award winner Prof. Yoshua Bengio.
Research Interests
- Architecture Research & Natural Language Processing — developing state-of-the-art neural network architectures, with applications to LLMs such as Gemini 2.5.
- Robustness & Generalization — novel regularization and data augmentation methods, including mixup, to improve generalization and mitigate memorization in deep networks.
- Optimization & Deep Learning Theory — understanding the loss surface of neural networks and non-convex optimization.
Honors & Awards
- Best Paper Award — IEEE Signal Processing Society (SPS), 2020
- Best Paper Honorable Mention — Association for Computational Linguistics (ACL), 2018
- Best Paper Honorable Mention — Neural Information Processing Systems (NeurIPS), 2011
- First Place — Unsupervised Transfer Learning Challenge (Phase 2), Pascal2
- First Place — Emotion Recognition in the Wild Challenge, 2013