Teaching machines to|

Research engineer at Google Brain. Creating curious machines able to explore the world. Interested in language models, discrete structures, and the limits of computation.


QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
Improve question answering with attention and clever data augmentation.
Adams Wei Yu, David Dohan, Minh-Thang Luong, Rui Zhao, Kai Chen, Mohammad Norouzi, Quoc V. Le

Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks
Transfer learn from 3D renderings to real world data.
Konstantinos Bousmalis, Nathan Silberman, David Dohan, Dumitru Erhan, Dilip Krishnan

Learning Hierarchical Semantic Segmentations of LIDAR Data
Create semantic segmentations from point cloud data.
David Dohan, Brian Matejek, Thomas Funkhouser

Get in touch @dmdohan · github · about