Geng Ji

Geng Ji is a research scientist at Facebook AI since 2019. His research interests include reinforcement learning, counterfactual reasoning, large-scale Bayesian inference and their applications in Facebook products ranging from recommendation systems to text understanding. He got his PhD in computer science from UC Irvine, and finished his undergrad in engineering physics in Tsinghua University, China.

Geng's Publications

July 18, 2021

CORE MACHINE LEARNING

Marginalized Stochastic Natural Gradients for Black-Box Variational Inference

Black-box variational inference algorithms use stochastic sampling to analyze diverse statistical models, like those expressed in probabilistic programming languages …

Geng Ji, Debora Sujono, Erik B. Sudderth

July 18, 2021

June 30, 2019

RESEARCH

NLP

Variational Training for Large-Scale Noisy-OR Bayesian Networks

We propose a stochastic variational inference algorithm for training large-scale Bayesian networks, where noisy-OR conditional distributions are used to capture higher-order relationships. One application is to the learning of hierarchical…

Geng Ji, Dehua Cheng, Huazhong Ning, Changhe Yuan, Hanning Zhou, Liang Xiong, Erik B. Sudderth

June 30, 2019