Query rewriting via large language models

J Liu, B Mozafari - ar** with poorly written queries
before passing them down to the query optimizer. Manual rewriting is not scalable, as it is …

Causal representation learning from multiple distributions: A general setting

K Zhang, S **e, I Ng, Y Zheng - arxiv preprint arxiv:2402.05052, 2024 - arxiv.org
In many problems, the measured variables (eg, image pixels) are just mathematical
functions of the latent causal variables (eg, the underlying concepts or objects). For the …

Score-based causal discovery of latent variable causal models

I Ng, X Dong, H Dai, B Huang, P Spirtes… - Forty-first International …, 2024 - openreview.net
Identifying latent variables and the causal structure involving them is essential across
various scientific fields. While many existing works fall under the category of constraint …

Learning Discrete Concepts in Latent Hierarchical Models

L Kong, G Chen, B Huang, EP **ng, Y Chi… - arxiv preprint arxiv …, 2024 - arxiv.org
Learning concepts from natural high-dimensional data (eg, images) holds potential in
building human-aligned and interpretable machine learning models. Despite its …

Adversarial-Causal Representation Learning Networks for Machine fault diagnosis under unseen conditions based on vibration and acoustic signals

F Wu, Z **ang, D **ao, Y Hao, Y Qin, H Pu… - … Applications of Artificial …, 2025 - Elsevier
To address the challenges of obtaining diverse data, domain generalization (DG) methods
for fault diagnosis have been developed. Domain adversarial methods are currently the …

Differentiable Causal Discovery For Latent Hierarchical Causal Models

P Prashant, I Ng, K Zhang, B Huang - arxiv preprint arxiv:2411.19556, 2024 - arxiv.org
Discovering causal structures with latent variables from observational data is a fundamental
challenge in causal discovery. Existing methods often rely on constraint-based, iterative …

On the Parameter Identifiability of Partially Observed Linear Causal Models

X Dong, I Ng, B Huang, Y Sun, S **, R Legaspi… - arxiv preprint arxiv …, 2024 - arxiv.org
Linear causal models are important tools for modeling causal dependencies and yet in
practice, only a subset of the variables can be observed. In this paper, we examine the …

Identifying Latent State-Transition Processes for Individualized Reinforcement Learning

Y Sun, B Huang, Y Yao, D Zeng… - Advances in …, 2025 - proceedings.neurips.cc
The application of reinforcement learning (RL) involving interactions with individuals has
grown significantly in recent years. These interactions, influenced by factors such as …

A Skewness-Based Criterion for Addressing Heteroscedastic Noise in Causal Discovery

Y Lin, Y Huang, W Liu, H Deng, I Ng, K Zhang… - arxiv preprint arxiv …, 2024 - arxiv.org
Real-world data often violates the equal-variance assumption (homoscedasticity), making it
essential to account for heteroscedastic noise in causal discovery. In this work, we explore …

Permutation-Based Rank Test in the Presence of Discretization and Application in Causal Discovery with Mixed Data

X Dong, I Ng, B Sun, H Dai, GY Hao, S Fan… - arxiv preprint arxiv …, 2025 - arxiv.org
Recent advances have shown that statistical tests for the rank of cross-covariance matrices
play an important role in causal discovery. These rank tests include partial correlation tests …