NNSmith: Generating Diverse and Valid Test Cases for Deep Learning Compilers

1 code implementation • 26 Jul 2022 • Jiawei Liu, JinKun Lin, Fabian Ruffy, Cheng Tan, Jinyang Li, Aurojit Panda, Lingming Zhang

In this work, we propose a new fuzz testing approach for finding bugs in deep-learning compilers.

Measuring the Effect of Training Data on Deep Learning Predictions via Randomized Experiments

1 code implementation • 20 Jun 2022 • JinKun Lin, Anqi Zhang, Mathias Lecuyer, Jinyang Li, Aurojit Panda, Siddhartha Sen

Our algorithm estimates the AME, a quantity that measures the expected (average) marginal effect of adding a data point to a subset of the training data, sampled from a given distribution.

Causal Inference

Hop: Heterogeneity-Aware Decentralized Training

no code implementations • 4 Feb 2019 • Qinyi Luo, JinKun Lin, Youwei Zhuo, Xuehai Qian

Based on a unique characteristic of decentralized training that we have identified, the iteration gap, we propose a queue-based synchronization mechanism that can efficiently implement backup workers and bounded staleness in the decentralized setting.