Publications

Regret-Based ($\epsilon, \delta$)-optimal Stopping Criteria for Bayesian Optimization

We propose theoretically sound stopping criteria for GP-UCB based on tighter instantaneous regret bounds, ensuring an $\epsilon$-optimal solution with high probability upon termination.

Haowei Wang, Jingyi Wang, Qiyu Wei 魏启宇

Counterfactual Credit Guided Bayesian Optimization

We introduce CCGBO, a Bayesian Optimization framework that uses counterfactual credit to quantify historical observations’ contribution to discovering the optimum, accelerating convergence.

Qiyu Wei 魏启宇, Haowei Wang, Richard Allmendinger, Mauricio A. Álvarez

Counterfactual Credit Guided Bayesian Optimization

THCM-CAL: Temporal-Hierarchical Causal Modelling with Conformal Calibration for Clinical Risk Prediction

We propose THCM-CAL, a temporal-hierarchical causal modeling framework with conformal calibration for clinical risk prediction from structured and unstructured electronic health records.

Xin Zhang, Qiyu Wei 魏启宇, Yingjie Zhu, Fanyi Wu, Sophia Ananiadou

SynGraph: A Dynamic Graph-LLM Synthesis Framework for Sparse Streaming User Sentiment Modeling

We propose SynGraph, a dynamic Graph-LLM synthesis framework that mitigates data sparsity in streaming sentiment analysis using Continuous-Time Dynamic Graphs and LLM-augmented representations.

Xin Zhang, Qiyu Wei 魏启宇, Yingjie Zhu, Linhai Zhang, Deyu Zhou, Sophia Ananiadou

SynGraph: A Dynamic Graph-LLM Synthesis Framework for Sparse Streaming User Sentiment Modeling

Gradient-based Sample Selection for Faster Bayesian Optimization

We propose GSSBO, a gradient-based sample selection method that scales Bayesian Optimization to large-budget scenarios by selecting a representative subset of historical observations based on gradient information.

Qiyu Wei 魏启宇, Haowei Wang, Zirui Cao, Songhao Wang, Richard Allmendinger, Mauricio A. Álvarez

Controlling Facial Attribute Synthesis by Disentangling Attribute Feature Axes in Latent Space

Face Image Synthesis, Face Attribute, Latent Vector, Gram-Schmidt orthogonalization.

Qiyu Wei 魏启宇, Weihua Zheng, Yuxin Li, Zhongyao Cheng, Zeng Zeng, Xulei Yang

Controlling Facial Attribute Synthesis by Disentangling Attribute Feature Axes in Latent Space

A Deeply Supervised Semantic Segmentation Method Based on GAN

LIn recent years, the field of intelligent transportation has witnessed rapid advancements, driven by the increasing demand for …

Wei Zhao, Qiyu Wei 魏启宇, Zeng Zeng

A Deeply Supervised Semantic Segmentation Method Based on GAN

Wafer Map Defect Patterns Semi-Supervised Classification Using Latent Vector Representation

Defect detection; semi-supervised learning; unsupervised learning; variation autoencoder

Qiyu Wei 魏启宇, Wei Zhao, Xiaoyan Zheng, Zeng Zeng

Bayesian optimization with switching cost: Regret analysis and lookahead variants

Machine Learning, Bayesian learning, Hyperparameter optimization

Peng Liu, Haowei Wang, Qiyu Wei 魏启宇

Bayesian optimization with switching cost: Regret analysis and lookahead variants

Latent Vector Prototypes Guided Conditional Face Synthesis

Conditional Face Generation, Generative Adversarial Networks, Prototype Clustering

Qiyu Wei 魏启宇, Xulei Yang$, Tong Sang$, Huijiao Wang, Zou Xiaofeng, Cheng Zhongyao, Zhao Ziyuan, Zeng Zeng

A job scheduling algorithm based on parallel workload prediction on computational grid

Generally, the computational grid consists of a large number of computing nodes, some of them are idle due to the uneven geographical …

Xiaoyong Tang, Yi Liu, Tan Deng, Zexin Zeng, Haowei Huang, Qiyu Wei 魏启宇, Xiaorong Li, Li Yang

A job scheduling algorithm based on parallel workload prediction on computational grid

Towards Efficient Point Cloud Classification via Critical Point Aware Pooling with Adaptive Learnable Threshold

Critical point aware pooling, Graph neural networks, Point clouds, Learnable importance.

Tong Sang, Qiyu Wei 魏启宇, Xiaofeng Zou, Zeng Zeng, Xulei Yang