Static Prediction of Runtime Errors by Learning to Execute Programs with External Resource Descriptions

1 code implementation • 7 Mar 2022 • David Bieber, Rishab Goel, Daniel Zheng, Hugo Larochelle, Daniel Tarlow

This presents an interesting machine learning challenge: can we predict runtime errors in a "static" setting, where program execution is not possible?

Learning to Execute

Swift for TensorFlow: A portable, flexible platform for deep learning

no code implementations • 26 Feb 2021 • Brennan Saeta, Denys Shabalin, Marc Rasi, Brad Larson, Xihui Wu, Parker Schuh, Michelle Casbon, Daniel Zheng, Saleem Abdulrasool, Aleksandr Efremov, Dave Abrahams, Chris Lattner, Richard Wei

Swift for TensorFlow is a deep learning platform that scales from mobile devices to clusters of hardware accelerators in data centers.

Learning Dexterous Manipulation from Suboptimal Experts

no code implementations • 16 Oct 2020 • Rae Jeong, Jost Tobias Springenberg, Jackie Kay, Daniel Zheng, Yuxiang Zhou, Alexandre Galashov, Nicolas Heess, Francesco Nori

Although in many cases the learning process could be guided by demonstrations or other suboptimal experts, current RL algorithms for continuous action spaces often fail to effectively utilize combinations of highly off-policy expert data and on-policy exploration data.

Offline RL Q-Learning

Demystifying Differentiable Programming: Shift/Reset the Penultimate Backpropagator

1 code implementation • 27 Mar 2018 • Fei Wang, Daniel Zheng, James Decker, Xilun Wu, Grégory M. Essertel, Tiark Rompf

Deep learning has seen tremendous success over the past decade in computer vision, machine translation, and gameplay.

Machine Translation Translation