About

I am a research scientist at Borealis AI working in Machine Learning for finance.

My interests lie in the intersection of 3D object understanding and robotics. This includes 3D object generation, and prediction from a variety of input modalities and 3D understanding for robot-object interaction.

Recent Activity
Papers
Scott Fujimoto, Edward J. Smith, Wei-Di Chang, Shixiang Shane Gu, Doina Precup, David Meger
For SALE: State-Action Representation Learning for Deep Reinforcement Learning
Edward J. Smith, Michal Drozdzal, Derek Nowrouzezahrai, David Meger, Adriana Romero-Soriano
Uncertainty-Driven Active Vision for Implicit Scene Reconstruction
Edward J. Smith, David Meger, Luis Pineda, Roberto Calandra, Jitendra Malik, Adriana Romero, Michal Drozdzal
Active 3D Shape Reconstruction from Vision and Touch
Conference on Neural Information Processing Systems (NeurIPS), 2021
Edward J. Smith, Roberto Calandra, Adriana Romero, Georgia Gkioxari, David Meger, Jitendra Malik, Michal Drozdzal
3D Shape Reconstruction from Vision and Touch
Conference on Neural Information Processing Systems (NeurIPS), 2020
Edward J. Smith, Krishna Murthy Jatavallabhula (Equal-first), Jean-Francois Lafleche, Clement Fuji Tsang, Artem Rozantsev, Wenzheng Chen, Tommy Xiang, Rev Lebaredian, Sanja Fidler
Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research
arXiv, 2019
Wenzheng Chen, Edward J. Smith*, Jun Gao*, Huan Ling*, Jaakko Lehtinen, Alec Jacobson, Sanja Fidler
Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer
Conference on Neural Information Processing Systems (NeurIPS), 2019
Edward J. Smith, Scott Fujimoto, Adriana Romero, David Meger
GEOMetrics: Exploiting Geometric Structure for Graph-Encoded Objects
International Conference on Machine Learning (ICML), 2019
Edward J. Smith, Scott Fujimoto, David Meger
Multi-View Silhouette and Depth Decomposition for High Resolution 3D Object Representation
Conference on Neural Information Processing Systems (NIPS), 2018
Edward J. Smith, David Meger
Improved Adversarial Systems for 3D Object Generation and Reconstruction
Conference on Robot Learning (CoRL), 2017