Publications by authors named "Debo Cheng"
Article Synopsis
- Graph Neural Networks (GNNs) effectively learn from graph-structured data but can unintentionally lead to discrimination against certain groups due to correlations with sensitive attributes.
- To address this issue, the Fair Disentangled Graph Neural Network (FDGNN) framework is introduced, which focuses on creating fair node representations by augmenting data diversity and addressing sensitive value distributions.
- Extensive experiments show that FDGNN outperforms traditional methods in fairness while demonstrating the effectiveness of disentanglement in learning unbiased representations.
Predicting drug-target interactions (DTI) is a crucial stage in drug discovery and development. Understanding the interaction between drugs and targets is essential for pinpointing the specific relationship between drug molecules and targets, akin to solving a link prediction problem using information technology. While knowledge graph (KG) and knowledge graph embedding (KGE) methods have been rapid advancements and demonstrated impressive performance in drug discovery, they often lack authenticity and accuracy in identifying DTI.
View Article and Find Full Text PDFDiscovering Ancestral Instrumental Variables for Causal Inference From Observational Data.
IEEE Trans Neural Netw Learn Syst
August 2024
Article Synopsis
- Instrumental variable (IV) methods are crucial for understanding causal effects from observational data, especially in the presence of hidden confounders, but selecting a valid IV is essential to avoid bias.
- This article introduces a data-driven algorithm designed to identify valid IVs by leveraging partial ancestral graphs (PAGs) and determining candidate ancestral IVs (AIVs) along with their conditioning sets.
- Experimental results demonstrate that this new IV discovery algorithm produces more accurate causal effect estimates than existing state-of-the-art IV methods when applied to both synthetic and real-world datasets.
Toward Unique and Unbiased Causal Effect Estimation From Data With Hidden Variables.
IEEE Trans Neural Netw Learn Syst
September 2023
Causal effect estimation from observational data is a crucial but challenging task. Currently, only a limited number of data-driven causal effect estimation methods are available. These methods either provide only a bound estimation of causal effects of treatment on the outcome or generate a unique estimation of the causal effect but making strong assumptions on data and having low efficiency.
View Article and Find Full Text PDF