Chao Zhang

My research aims to build intelligent systems that help people better make use of their data for task support and decision making, especially text data. I am particularly interested in making intelligent systems more label-efficient and robust in open-world settings. Towards this goal, I and my awesome students are currently working on the following research thrusts:

• Knowledge extraction from text: Most of human knowledge is encoded in text. How can we extract from unstructured text structured knowledge (e.g., entities, taxonomies, events) that can be used by machines? How do we extract knowledge for domain-specific corpora that are low-resource in nature? And how can we use the knowledge to facilitate tasks such as scientific discovery?
  • STEAM: Self-Supervised Taxonomy Expansion with Mini-Paths, KDD 2020
  • BOND: Bert-Assisted Open-Domain Named Entity Recognition with Distant Supervision, KDD 2020
  • SeqMix: Augmenting Active Sequence Labeling via Sequence Mixup, EMNLP 2020
  • TaxoGen: Unsupervised Topic Taxonomy Construction by Adaptive Term Embedding and Clustering, KDD 2018
  • TrioVecEvent: Embedding-Based Online Local Event Detection in Geo-Tagged Tweet Streams, KDD 2017
  • GeoBurst: Real-Time Local Event Detection in Geo-Tagged Tweet Streams, SIGIR 2016
• Uncertainty and open-world learning: Deep learning models can be poor at estimating their uncertainties and fail miserably in the open domain. How can we quantify the uncertainties for deep learning models? How can we make machine learning models generalize to out-of-distribution samples? How to make them say "I don't know" when facing unseen tasks and quickly adapt to them?
  • Calibrated Fine-Tuning for Pre-trained Language Models via Manifold Smoothing, EMNLP 2020
  • SDE-Net: Equipping Deep Neural Networks with Uncertainty Estimates, ICML 2020
  • RDeepSense: Reliable Deep Mobile Computing Models with Uncertainty Estimations, UbiComp 2018
• Weakly-supervised learning on text data: A major bottleneck of deep learning models is their label inefficiency. When there are no labeled data but just some weak supervision converted from domain knowledge, can we still use such weak supervision to train effective models? How do we handle the label noise and label coverage issues when learning from such weak supervision?
  • Text Classification Using Label Names Only: A Language Model Self-Training Approach, EMNLP 2020
  • Denoising Multi-Source Weak Supervision for Neural Text Classification, ENNLP 2020
  • Joint Aspect-Sentiment Analysis with Minimal User Guidance, SIGIR 2020
  • Doc2Cube: Allocating Documents to Text Cube without Labeled Data, ICDM 2018
  • Weakly-Supervised Neural Text Classification, CIKM 2018
• Composable and interpretable models: When we know the structure of a problem, can we leverage that to design more effective models? When some modules are shared across tasks, how can we effectively reuse such modules without suffering from too strong assumptions?
  • State-Sharing Sparse Hidden Markov Models for Personalized Sequences, KDD 2019
  • GMove: Group-Level Mobility Modeling Using Geo-Tagged Social Media, KDD 2016

• We have 4 paper accepted by EMNLP'20, discussing model calibration, sequence labeling, and zero-shot text classification.
• We have 1 paper accepted by ICML'20. We introduce a new method for quantifying uncertainties for neural networks based on the connection between NN and linear systems.
• We have 4 papers accepted by KDD'20, discussing self-supervised taxonomy construction, low-resource NER, hierarchical topic mining, and tensor factorization.
• Congrats to my student Wendi Ren for winning the Marshall D. Williamson Fellowship!
• Honored to receive the 2020 Google Faculty Research Award!
• Two papers accepted by the Web Conference 2020.
• Honored to receive the ACM SIGKDD 2019 Dissertation Runner-up Award!
• Our monograph Multidimensional Mining of Massive Text Data is published by Morgan & Claypool!

• 2020 Amazon AWS Machine Learning Research Award
• 2020 Google Faculty Research Award
• 2019 ACM SIGKDD Dissertation Award Runner-up
• 2018 ACM IMWUT Distinguished Paper Award
• 2015 ECML/PKDD Best Student Paper Runner-up Award
• 2013 Chiang Chen Overseas Graduate Fellowship

(* denotes equal contribution)

• 2020 Fall: CSE8803-DLT: Deep Learning for Text Data
• 2020 Spring: CX4240: Introduction to Computational Data Analysis
• 2019 Fall: CSE8803-DLT: Deep Learning for Text Data
• 2019 Spring: CX4240: Introduction to Computational Data Analysis

• Ph.D. Students:
  • Lingkai Kong
  • Rui Feng
  • Yue Yu
  • Yinghao Li
  • Yi Rong (Visiting)
• Master Students:
  • Rongzhi Zhang
  • Yuchen Zhuang
  • Isaac Rehg
  • Wendi Ren
  • Ruijia Wang