Yujia Zheng (郑雹昉)

I am an MS student at CMU Philosophy, advised by Prof. Kun Zhang. I recently completed my BS from University of Electronic Science and Technology of China. I previously interned at UC Berkeley and NLPR, and was a summer fellow at EPFL.

My research interests lie primarily in causality and machine learning. Currently, I mainly study the following topics: causal discovery, causal representation learning, and structure learning. I am also interested in the intersection between causality and deep learning.

Email  /  Google Scholar  /  LinkedIn

profile photo
causal-learn causal-learn: Causal Discovery for Python

Documentation, GitHub

Causal-learn is a Python translation and extension of the Tetrad java code. It offers the implementations of up-to-date causal discovery methods as well as simple and intuitive APIs. Causal-learn is contributed by several groups and I am the major coordinator. This project is under active development, and any comments or suggestions are welcome.


arXiv On the Identifiability of Nonlinear ICA: Sparsity and Beyond
Yujia Zheng, Ignavier Ng, Kun Zhang
NeurIPS 2022, arXiv

We prove the identifiability of nonlinear ICA with assumptions only on the mixing process, such as Structural Sparsity.

ICML Partial Disentanglement for Domain Adaptation
Lingjing Kong, Shaoan Xie, Weiran Yao, Yujia Zheng, Guangyi Chen, Petar Stojanov, Victor Akinwande, Kun Zhang
ICML 2022, Link

We show that under reasonable assumptions on the data generating process, as well as leveraging the principle of minimality, we can obtain partial identifiability of the changing and invariant parts of the generating process.

ICLR On the Identifiability of Nonlinear ICA with Unconditional Priors
Yujia Zheng, Ignavier Ng, Kun Zhang
ICLR 2022 OSC, Oral, Link

We aim to show the identifiability of nonlinear ICA with unconditional priors under specific conditions on the mixing process, such as independent influences.

NeurIPS Reliable Causal Discovery with Improved Exact Search and Weaker Assumptions
Ignavier Ng, Yujia Zheng, Jiji Zhang, Kun Zhang
NeurIPS 2021, arXiv

We introduce several strategies to improve the scalability of exact score-based methods in the linear Gaussian setting with theoretical guarantees under weaker assumptions.

arXiv Source Free Unsupervised Graph Domain Adaptation
Haitao Mao, Lun Du, Yujia Zheng, Qiang Fu, Zelin Li, Xu Chen, Shi Han, Dongmei Zhang
Preprint, arXiv

We propose a new scenario named source free unsupervised graph domain adaptation. In this scenario, the only information we can leverage from the source domain is the well-trained source model, without any exposure to the source graph and its labels.

KDD Learning Elastic Embeddings for Customizing On-Device Recommenders
Tong Chen, Hongzhi Yin, Yujia Zheng, Zi Huang, Yang Wang, Meng Wang
KDD 2021, arXiv

We present a novel lightweight recommendation paradigm that allows a well-trained recommender to be customized for arbitrary device-specific memory constraints without retraining.

AAAI Cold-start Sequential Recommendation via Meta Learner
Yujia Zheng, Siyi Liu, Zekun Li, Shu Wu
AAAI 2021, arXiv

We explore meta-learning to alleviate cold-start problem without any kind of side information.

RecSys Long-tail Session-based Recommendation
Siyi Liu, Yujia Zheng (corresponding author)
RecSys 2020, arXiv

We introduce Long-tail Recommendation into Session-based Recommendation.

ICDM DGTN: Dual-channel Graph Transition Network for Session-based Recommendation
Yujia Zheng, Siyi Liu, Zekun Li, Shu Wu
ICDM 2020 NeuRec, Long Oral, arXiv

We model the cross-session transitions via a channel-aware graph neural network.

arXiv Heterogenous Graph Collaborative Filtering
Zekun Li*, Yujia Zheng*, Shu Wu, Xiaoyu Zhang, Liang Wang
Preprint, arXiv

We model user-item interactions as a heterogeneous graph which consists of various edge types.

arXiv Balancing Multi-level Interactions for Session-based Recommendation
Yujia Zheng* (corresponding author), Siyi Liu*, Zailei Zhou
Preprint, arXiv

We introduce Inter-session Item-level Interactions into Session-based Recommendation.

ACAI ADS: Multimedia Dance Video Automatic Scoring Framework Based on Transfer Learning
Yujia Zheng
ACAI 2020, Link

We propose a multimedia dance video automatic scoring framework.

IGTA ReFall: Real-time Fall Detection of Continuous Depth Maps with RFD-Net
Yujia Zheng, Siyi Liu, Zairong Wang, Yunbo Rao
IGTA 2019, Link

We propose a real-time fall detection method for the elderly.


Aug. 2021 - Present , Carnegie Mellon University ,

MS, Logic, Computation & Methodology


Aug. 2017 - June. 2021 , University of Electronic Science and Technology of China ,

BS, Software Engineering, GPA: 3.98/4.0.


Aug. 2019 - Dec. 2019 , University of California, Berkeley ,

Visiting Student, Computational Approaches to Human Learning (CAHL) Research Lab, GPA: 4.0/4.0


Feb. 2019 - June. 2019 , National Taiwan University of Science and Technology ,

Exchange Student, Image/Video Processing and Multimedia Lab (IVPML), GPA: 4.0/4.0

Organizational Activities

Publicity Co-Chair of the 38th Conference on Uncertainty in Artificial Intelligence (UAI 2022).

Reviewer/Program Committee

Neural Information Processing Systems (NeurIPS).

International Conference on Machine Learning (ICML).

International Joint Conference on Artificial Intelligence (IJCAI).

AAAI Conference on Artificial Intelligence (AAAI).

The ACM Web Conference (WWW).

The ACM Web Search and Data Mining Conference (WSDM).

Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD).

The International AAAI Conference on Web and Social Media (ICWSM).

IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI).

IEEE Transactions on Big Data (TBD).


Member of youth delegation in the 73rd session of the UN General Assembly (UNGA 73).

Summer Research Fellowship at EPFL (<2%).

Top 1% in multiple data competitions.

Tang Lixin Scholarship.

Can make a living by playing Erhu on the street.

This awesome template is from here. Updated: 2022/05