Dilin Wang


The Artificial Intelligence Lab

Department of Computer Science

UT Austin

Email: dilin@cs.utexas.edu


I am a 5th year computer science Ph.D. student at University of Texas at Austin, advised by Professor Qiang Liu. My research interests are in statistical machine learning and deep learning.

[ Google Scholar ]  [ Github

Publications

Conference papers (* equal contribution)

Splitting Steepest Descent for Growing Neural Architectures

Qiang Liu, Lemeng Wu* and Dilin Wang*   

Neural Information Processing Systems (NeurIPS) 2019    Spotlight acceptance rate 2.4%

[paper]

Stein Variational Gradient Descent With Matrix-Valued Kernels

Dilin Wang*, Ziyang Tang*, Chandrajit Bajaj and Qiang Liu   

Neural Information Processing Systems (NeurIPS) 2019

[paper]    [code]

Mixed Precision Neural Architecture Search for Energy Efficient Deep Learning

Chengyue Gong*, Zixuan Jiang*, Dilin Wang, Yibo Lin, Qiang Liu, David Z. Pan   

IEEE/ACM International Conference on Computer-Aided Design (ICCAD) 2019

[paper]

Nonlinear Stein Variational Gradient Descent for Learning Diversified Mixture Models

Dilin Wang and Qiang Liu

International Conference on Machine Learning (ICML) 2019

[paper]    [code]

Improving Neural Language Modeling via Adversarial Training

Dilin Wang*, Chengyue Gong* and Qiang Liu   

International Conference on Machine Learning (ICML) 2019

[paper]    [code]

Variational Inference with Tail-adaptive f-Divergence

Dilin Wang, Hao Liu and Qiang Liu

Advances in Neural Information Processing Systems (NeurIPS) 2018   Oral acceptance rate 0.6%

[paper]    [code]    [slides]

Stein Variational Gradient Descent as Moment Matching

Qiang Liu and Dilin Wang

Advances in Neural Information Processing Systems (NIPS) 2018

[paper]   [code]

Stein Variational Message Passing for Continuous Graphical Models

Dilin Wang*, Zhe Zeng* and Qiang Liu   

International Conference on Machine Learning (ICML) 2018

[paper]    [code]

Learning to Draw Samples with Amortized Stein Variational Gradient Descent

Yihao Feng, Dilin Wang, Qiang Liu

The Conference on Uncertainty in Artificial Intelligence (UAI) 2017

[paper]    [code]

Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm

Qiang Liu and Dilin Wang

Advances in Neural Information Processing Systems (NIPS) 2016

[paper]    [code]

Entity Disambiguation by Knowledge and Text Jointly Embedding

Wei Fang, Jianwen Zhang, Dilin Wang, Zheng Chen and Ming Li

Conference on Natural Language Learning (CONLL) 2016

[paper]

Efficient Observation Selection in Probabilistic Graphical Models Using Bayesian Lower Bounds

Dilin Wang, John Fisher III and Qiang Liu

The Conference on Uncertainty in Artificial Intelligence (UAI) 2016   Oral

[paper]

Workshop papers

Energy-Aware Neural Architecture Optimization with Fast Splitting Steepest Descent

Dilin Wang, Meng Li, Lemeng Wu, Vikas Chandra and Qiang Liu

A short version will be presented at NeurIPS 2019 Workshop on Energy Efficient Machine Learning and Cognitive Computing (contributed talk)

[arxiv]    [code]

An Optimization View on Dynamic Routing Between Capsules

Dilin Wang and Qiang Liu

ICLR 2018 Workshop

[paper]

Learning to Draw Samples: With Application to Amortized MLE for Generative Adversarial Learning

Dilin Wang and Qiang Liu

A short version was presented at NIPS Bayesian Deep Learning Workshop (contributed talk) 2016

[paper]     [code]

Experiences

•  2019 summer  Research Intern at Facebook Research, Menlo Park, CA

•  2016 summer  Research Intern at Microsoft Research Redmond, WA

•  2013.11 - 2014.08  Research Intern at Microsoft Research Asia, Beijing

Service

I am/was a reviewer for: ICML'18, ICML'19 (top 5% reviewer), ICML'20, NIPS'18, NeurIPS'19, ICLR'19, ICLR'20, CVPR'19, ICCV'19, ECCV'20