Author name cluster

Shui Yu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

AAAI Conference 2026 Conference Paper

Priority-Based Graph-Enhanced Reinforcement Learning for Robust Analog Circuit Optimization

Jintao Li
Zhenxin Chen
Sicheng He
Ao-Jin Li
Shui Yu

A primary motivation for analog integrated circuit (IC) design automation is the inefficiency of manual design in meeting increasingly stringent specifications, which often involve over 10 objectives. Recent advances in reinforcement learning (RL) emerge as a promising method, yet gaps remain when considering full design specifications, especially under process-voltage-temperature (PVT) variations. Excessive objectives lead to diminished reward signals, while varying PVT conditions result in conflicting gradients, both of which result in inefficient exploration. To address these, we propose a priority-based graph-enhanced RL framework. Specifically, using fuzzy logic converts quantitative rewards into qualitative priority signals, mitigating reward deterioration and enhancing exploration via entropy regularization. Furthermore, a graph-based representation compresses high-dimensional objective spaces under PVT variations into low-dimensional manifolds, enabling dynamic resource allocation to variation-sensitive regions and resolving gradient conflicts. Empirical results on various real-world analog ICs demonstrate that our method significantly outperforms existing RL algorithms, achieving superior solution quality and reducing simulation overhead.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Revealing Multimodal Causality with Large Language Models

Jin Li
Shoujin Wang
Qi Zhang
Feng Liu
Tongliang Liu
Longbing Cao
Shui Yu
Fang Chen

Uncovering cause-and-effect mechanisms from data is fundamental to scientific progress. While large language models (LLMs) show promise for enhancing causal discovery (CD) from unstructured data, their application to the increasingly prevalent multimodal setting remains a critical challenge. Even with the advent of multimodal LLMs (MLLMs), their efficacy in multimodal CD is hindered by two primary limitations: (1) difficulty in exploring intra- and inter-modal interactions for comprehensive causal variable identification; and (2) insufficiency to handle structural ambiguities with purely observational data. To address these challenges, we propose MLLM-CD, a novel framework for multimodal causal discovery from unstructured data. It consists of three key components: (1) a novel contrastive factor discovery module to identify genuine multimodal factors based on the interactions explored from contrastive sample pairs; (2) a statistical causal structure discovery module to infer causal relationships among discovered factors; and (3) an iterative multimodal counterfactual reasoning module to refine the discovery outcomes iteratively by incorporating the world knowledge and reasoning capabilities of MLLMs. Extensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of the proposed MLLM-CD in revealing genuine factors and causal relationships among them from multimodal unstructured data. The implementation code and data are available at https: //github. com/JinLi-i/MLLM-CD.

PDF Details

EAAI Journal 2024 Journal Article

Game-theoretic analytics for privacy preservation in Internet of Things networks: A survey

Yizhou Shen
Carlton Shepherd
Chuadhry Mujeeb Ahmed
Shigen Shen
Xiaoping Wu
Wenlong Ke
Shui Yu

Privacy preservation of the big data generated, deposited, and communicated by smart IoT (Internet of Things) nodes is the major challenge in IoT networks. Anonymization, encryption, and routing protocol constitute the existing prevalent privacy-preserving approaches, most of which have successfully implemented the privacy preservation of data query, data mining and data aggregation. Nevertheless, there has been a gradual switch in the selection of privacy-preserving technology. Predictive game-theoretic analytics for privacy preservation in IoT networks has received significant attention since it can effectively settle the conflicts between attackers and defenders. In this survey, we explain the basics of various games mainly applied for IoT privacy preservation, such as simultaneous game, stochastic game, bargain game, differential game, mean field game, aggregation game, Stackelberg game, signaling game, repeated game, evolutionary game, and cooperative game. We then explore different applications for game theory-based privacy preservation in IoT networks, followed by discussing the differences among the existing solution of privacy-preserving issues using different games under specific IoT scenarios. Moreover, we consider the challenges and outline future research directions. In conclusion, this survey not only presents existing work on applying game theory to preserve privacy in current IoT networks including smart grids, intelligent transportation systems, crowdsensing, edge-based IoT, integrated energy systems, blockchain IoT, Social IoT and Industrial IoT, but it also encourages researches to further dig deeper into rare areas.

Details DOI

IS Journal 2023 Journal Article

The SRVM: A Similarity-Based Relevance Vector Machine for Remaining Useful Lifetime Prediction in the Industrial Internet of Things

Guorui Li
Yajun Wu
Cong Wang
Sancheng Peng
Jianwei Niu
Shui Yu

With the continuous advancement of Industry 4. 0 and intelligent manufacturing, remaining useful lifetime (RUL) prediction can forecast the future degradation state of machinery and then estimate the remaining service time before it loses its safe operation ability. Accordingly, a series of predictive maintenance strategies can be regulated in advance for equipment in the Industrial Internet of Things. To tackle the challenges of insufficiency of failure data and lack of confidence in RUL prediction results, a similarity-based relevance vector machine (SRVM) is proposed in this article. Primarily, the relationship among latent variables in the SRVM is learned adaptively through similarity computations to fully utilize the limited degradation data. Furthermore, these internal variables in the SRVM are treated as time-varying variables and re-estimated dynamically to provide RUL prediction with reliable confidence. The experiment results show that the prediction accuracy of the SRVM is higher than that of other baseline methods.

Details DOI

EAAI Journal 2022 Journal Article

Value-based reinforcement learning approaches for task offloading in Delay Constrained Vehicular Edge Computing

Do Bao Son
Ta Huu Binh
Hiep Khac Vo
Binh Minh Nguyen
Huynh Thi Thanh Binh
Shui Yu

In the age of booming information technology, human-being has witnessed the need for new paradigms with both high computational capability and low latency. A potential solution is Vehicular Edge Computing (VEC). Previous work proposed a Fuzzy Deep Q-Network in Offloading scheme (FDQO) that combines Fuzzy rules and Deep Q-Network (DQN) to improve DQN’s early performance by using Fuzzy Controller (FC). However, we notice that frequent usage of FC can hinder the future growth performance of model. One way to overcome this issue is to remove Fuzzy Controller entirely. We introduced an algorithm called baseline DQN (b-DQN), represented by its two variants Static baseline DQN (Sb-DQN) and Dynamic baseline DQN (Db-DQN), to modify the exploration rate base on the average rewards of closest observations. Our findings confirm that these baseline DQN algorithms surpass traditional DQN models in terms of average Quality of Experience (QoE) in 100 time slots by about 6%, but still suffer from poor early performance (such as in the first 5 time slots). Here, we introduce baseline FDQO (b-FDQO). This algorithm has a strategy to modify the Fuzzy Logic usage instead of removing it entirely while still observing the rewards to modify the exploration rate. It brings a higher average QoE in the first 5 time slots compared to other non-fuzzy-logic algorithms by at least 55. 12%, prevent the model from getting too bad result over all time slots, while having the late performance as good as that of b-DQN.

Details DOI