Yuhui Shi

Southern University of Science and Technology | SUSTech · Department of Computer Science and Engineering

Traditional multiobjective optimization problems (MOPs) are insufficiently equipped for scenarios involving multiple decision makers (DMs), which are prevalent in many practical applications. These scenarios are categorized as multiparty multiobjective optimization problems (MPMOPs). For MPMOPs, the goal is to find a solution set that is as close t...

Large Language Models (LLMs) have made significant strides in the field of artificial intelligence, showcasing their ability to interact with humans and influence human cognition through information dissemination. However, recent studies have brought to light instances of bias inherent within these LLMs, presenting a critical issue that demands att...

In the post-Moore era, main performance gains of black-box optimizers are increasingly depending on parallelism, especially for large-scale optimization (LSO). Here we propose to parallelize the well-established covariance matrix adaptation evolution strategy (CMA-ES) and in particular its one latest LSO variant called limited-memory CMA-ES (LM-CMA...

Decentralized collaborative learning for Point-of-Interest (POI) recommendation has gained research interest due to its advantages in privacy preservation and efficiency, as it keeps data locally and leverages collaborative learning among clients to train models in a decentralized manner. However, since local data is often limited and insufficient...

Gradient inversion attacks (GIAs) have posed significant challenges to the emerging paradigm of distributed learning, which aims to reconstruct the private training data of clients (participating parties in distributed training) through the shared parameters. For counteracting GIAs, a large number of privacy-preserving methods for distributed learn...

Deep Neural Networks (DNNs) have been widely used in many areas such as autonomous driving and face recognition. However, DNN model is fragile to backdoor attack. A backdoor in the DNN model can be activated by a poisoned input with trigger and leads to wrong prediction, which causes serious security issues in applications. It is challenging for cu...

To address privacy concerns and reduce network latency, there has been a recent trend of compressing cumbersome recommendation models trained on the cloud and deploying compact recommender models to resource-limited devices for the real-time recommendation. Existing solutions generally overlook device heterogeneity and user heterogeneity. They requ...

Influence maximization aims at finding a set of nodes with the maximum influence in a network. Apart from the influence, the burden of nodes is worth considering as an objective but has been neglected in the literature. Herein, the burden refers to the capacity of a node to serve its influenced nodes, e.g., the storage of a device in the device-to-...

Automated design of metaheuristic algorithms offers an attractive avenue to reduce human effort and gain enhanced performance beyond human intuition. Current automated methods design algorithms within a fixed structure and operate from scratch. This poses a clear gap towards fully discovering potentials over the metaheuristic family and fertilizing...

Large models involve solving more complex problems using intelligent computing methods, such as machine learning, evolutionary algorithms, and swarm intelligence. Multiple choices are a common phenomenon in consumer electronics. Multimodal optimization (MMO) aims to find multiple satisfied solutions simultaneously. The number (quantity) and the acc...

Gradient inversion attacks (GIAs) pose significant challenges to the privacy-preserving paradigm of distributed learning. These attacks employ carefully designed strategies to reconstruct victim’s private training data from their shared gradients. However, existing work mainly focuses on attacks and defenses for image-modal data, while the study fo...

Multi-agent reinforcement learning (MARL) has achieved encouraging performance in solving complex multi-agent tasks. However, the safety of MARL policies is one critical concern that impedes their real-world applications. Furthermore, popular multi-agent benchmarks provide limited safety support for safe MARL research, where negative rewards for co...

Some evolutionary algorithms have been proposed to address biparty multiobjective optimization problems (BPMOPs). However, all these algorithms are centralized algorithms which directly obtain the privacy information including objective functions from decision makers (DMs). This paper transforms the centralized algorithm OptMPNDS2 into a distribute...

To address privacy concerns and reduce network latency, there has been a recent trend of compressing cumbersome recommendation models trained on the cloud and deploying compact recommender models to resource-limited devices for real-time recommendation. Existing solutions generally overlook device heterogeneity and user heterogeneity. They either r...

Smoothing rugged and rough fitness landscapes by machine learning (ML) models is a feasible approach to improve the global optimization performance of evolutionary algorithms (EA). However, systematic studies on the smoothing performance of ML models are rare. Moreover, existing model management methods have a relatively low efficiency and can caus...

Given the ubiquity of non-separable optimization problems in real worlds, in this paper we analyze and extend the large-scale version of the well-known cooperative coevolution (CC), a divide-and-conquer optimization framework, on non-separable functions. First, we reveal empirical reasons of why decomposition-based methods are preferred or not in p...

Collaborative filtering (CF) based recommender systems are typically trained based on personal interaction data (e.g., clicks and purchases) that could be naturally represented as ego graphs. However, most existing recommendation methods collect these ego graphs from all users to compose a global graph to obtain high-order collaborative information...

Recommender systems play a significant role in information filtering and have been utilized in different scenarios, such as e-commerce and social media. With the prosperity of deep learning, deep recommender systems show superior performance by capturing non-linear information and item-user relationships. However, the design of deep recommender sys...

Model inversion attacks aim to reveal information about sensitive training data of AI models, which may lead to serious privacy leakage. However, existing attack methods have limitations in reconstructing training data with higher feature fidelity. In this paper, we propose an evolutionary model inversion attack approach (EvoMI) and empirically dem...

In this paper, we present a pure-Python open-source library, called PyPop7, for black-box optimization (BBO). It provides a unified and modular interface for more than 60 versions and variants of different black-box optimization algorithms, particularly population-based optimizers, which can be classified into 12 popular families: Evolution Strateg...

PPT

The increase of computing power can be continuously driven by parallelism, despite of the end of Moore’s law. To cater to this trend, we propose to parallelize the low-memory matrix adaptation evolution strategy (LM-MA-ES) recently proposed for large-scale black-box optimization, aiming at further improving its scalability (w.r.t. CPU cores) in the...

In the field of evolutionary multiobjective optimization, the decision maker (DM) concerns conflicting objectives. In the real-world applications, there usually exist more than one DM and each DM concerns parts of these objectives. Multiparty multiobjective optimization problems (MPMOPs) are proposed to depict the MOP with multiple decision makers...

As a special class of multiobjective optimization problems (MOPs), multiparty multiobjective optimization problems (MPMOPs) widely exist in real-world applications. In MPMOPs, there are multiple decision makers (DMs) concerning multiple different conflicting objectives. The goal of solving MPMOPs is to catch the best solutions satisfying all DMs as...

The multiple-target self-organizing pursuit (SOP) problem has wide applications and has been considered a challenging self-organization game for distributed systems, in which intelligent agents cooperatively pursue multiple dynamic targets with partial observations. This work proposes a framework for decentralized multi-agent systems to improve int...

Recommender systems play a significant role in information filtering and have been utilised in different scenarios, such as e-commerce and social media. With the prosperity of deep learning, deep recommender systems show superior performance by capturing non-linear information and item-user relationships. However, the design of deep recommender sys...

This paper presents a novel methodology to detect communities in complex networks based on evolutionary computation. In the proposed method, a surrogate network with a more detectable community structure than the original network is firstly constructed based on the eigenmatrix of the adjacent matrix. Then the community partition can be found by suc...

Offline reinforcement learning (RL) aims to train an agent solely using a dataset of historical interactions with the environments without any further costly or dangerous active exploration. Model-based RL (MbRL) usually achieves promising performance in offline RL due to its high sample-efficiency and compact modeling of a dynamic environment. How...

Dynamic and multimodal features are two important properties and widely existed in many real-world optimization problems. The former illustrates that the objectives and/or constraints of the problems change over time, while the latter means there is more than one optimal solution (sometimes including the accepted local solutions) in each environmen...

Brain storm optimization (BSO) groups population into several clusters and generates new individuals by using the information of these clusters. However, this mechanism limits the ability of exploration because it prevents new individuals from searching regions far away from current clusters. In this paper, we innovatively propose a grid-based sear...

Recently, covariance matrix adaption evolution strategy (CMA-ES) and its variants have achieved great success in the continuous unimodal optimization tasks owing to its strong local search capabilities. However, it is precisely this capability that reduces the population diversity, which makes it unable to obtain the good performance on the multimo...

Various niching methods have been widely adopted for solving multimodal optimization. However, keeping a balance between exploitation and exploration is still a tough task for designers of multimodal optimization algorithms. An essential niching method is encouraged to deal with optimization problems. In this paper, we proposed an adaptive niching...

Swarm intelligence optimization algorithms can be adopted in swarm robotics for target searching tasks in a 2-D or 3-D space by treating the target signal strength as fitness values. Many current works in the literature have achieved good performance in single-target search problems. However, when there are multiple targets in an environment to be...

Brain storm optimization (BSO) is an emerging global optimization algorithm. The primary idea is to divide the population into different clusters, and offspring are generated within a cluster or between two clusters. However, the problems of inefficient clustering strategy and insufficient exploration exist in BSO. In this paper, a novel and effici...

Imbalanced classification on graphs is ubiquitous yet challenging in many real-world applications, such as fraudulent node detection. Recently, graph neural networks (GNNs) have shown promising performance on many network analysis tasks. However, most existing GNNs have almost exclusively focused on the balanced networks, and would get unappealing...

Dynamic multi-objective optimization (DMO) has recently attracted increasing interest. Suitable benchmark problems are crucial for evaluating the performance of DMO solvers. However, most of the existing DMO benchmarks mainly focus on Pareto-optimal solutions (PS) varying on the hyperplane, which may produce some unexpected bias for algorithmic ana...

Automatic extracting of knowledge from massive data samples, i.e., big data analytics (BDA), has emerged as a vital task in almost all scientific research fields. The BDA problems are rather difficult to solve due to their large-scale, high-dimensional, and dynamic properties, while the problems with small data are usually hard to handle due to ins...

This paper proposes a novel homing behavior model of pigeons, and the simulation results are presented. A homing pigeon-inspired autonomous navigation system for unmanned aerial vehicle (UAV) is further achieved. Experimental results on the designed simulation platform show that UAV can finish an autonomous return voyage similar to the homing pigeo...

The evolutionary game theory aims to simulate different decision strategies in populations of individuals and to determine how the population evolves. Compared to strategies between two agents, such as cooperation or noncooperation, strategies on multiple agents are rather challenging and difficult to be simulated via traditional methods. Particula...

We can not see everything around us. Instead, the visual attention mechanism will select some fixations from extensive visual information for further processing. Many computational attention models have been proposed by imitating this mechanism. However, almost all of the state-of-the-art computational models output a complete saliency map, which m...

The Dynamic Vehicle Routing Problem (DVRP) has many real-world applications and practical values. The objective of DVRP is to find the optimal routes for a fleet of vehicles to service the given customer requests, without violating the vehicle capacity constraint. In this paper, a hybrid algorithm is proposed for solving the DVRP with the objective...

Many real-world problems involve cooperation and/or competition among multiple agents. These problems often can be formulated as multi-agent problems. Recently, Reinforcement Learning (RL) has made significant progress on single-agent problems. However, multi-agent problems still cannot be easily solved by traditional RL algorithms. First, the mult...

Brain storm optimization algorithms (BSO) have shown great potential in many global black-box optimization problems. However, the existing BSO variants can suffer from three problems: (1) large-scale optimization problem; (2) hyperparameter optimization problem; (3) high computational cost of the clustering operations. To address these problems, in...

The Swarm Intelligence (SI) algorithms have been proved to be a comprehensive method to solve complex optimization problems by simulating the emergence behaviors of biological swarms. Nowadays, data science is getting more and more attention, which needs quick management and analysis of massive data. Most traditional methods can only be applied to...

Parameter control is critical for the performance of any swarm intelligence algorithm. In this study, we propose an adaptive online data-driven closed-loop parameter control (CLPC) strategy for a swarm intelligence algorithm to solve both single-objective and multi-objective optimization problems with better performance. The proposed CLPC strategy...

This paper presents a limit-cycle-based mutant multi-objective pigeon-inspired optimization. In this algorithm, the limit-cycle-based mechanism is devised to consider the factors that affect the flight of pigeons to simplify the multi-objective pigeon-inspired optimization algorithm. The mutant mechanism is incorporated to strengthen the exploratio...

Due to the efficiency and efficacy in performance to tackle complex optimization problems, swarm intelligence (SI) optimizers, newly emerged as nature-inspired algorithms, have gained great interest from researchers over different fields. A large number of SI optimizers and their extensions have been developed, which drives the need to comprehensiv...

In brain storm optimization (BSO), the convergent operation utilizes a clustering strategy to group the population into multiple clusters, and the divergent operation uses this cluster information to generate new individuals. However, this mechanism is inefficient to regulate the exploration and exploitation search. This article first analyzes the...

The pursuit domain, or predator–prey problem is a standard testbed for the study of coordination techniques. In spite that its problem setup is apparently simple, it is challenging for the research of the emerged swarm intelligence. This paper presents a particle swarm optimization (PSO) based cooperative coevolutionary algorithm for the (predator)...

Evolutionary multi-task optimization (EMTO) is an emerging research topic in the field of evolutionary computation, which aims to simultaneously optimize several component tasks within a problem and output the best solution for each task. Since EMTO has widespread applications in solving real-world multi-task optimization problems, in recent years,...

Fireworks algorithms for solving problems with the optima shifts in the decision space and/or objective space are analyzed. The standard benchmark problems have several weaknesses in the research of swarm intelligence algorithms for solving single-objective problems. The optimum shift in decision space and/or objective space will increase the diffi...

Because of successful implementations and high intensity, metaheuristic research has been extensively reported in literature, which covers algorithms, applications, comparisons, and analysis. Though, little has been evidenced on insightful analysis of metaheuristic performance issues, and it is still a “black box” that why certain metaheuristics pe...

It is obvious from wider spectrum of successful applications that metaheuristic algorithms are potential solutions to hard optimization problems. Among such algorithms are swarm-based methods like particle swarm optimization and ant colony optimization which are increasingly attracting new researchers. Despite popularity, the core questions on perf...

The invention portfolio discloses a control method and system of a group robot. The method comprises that a prediction model is established on the basis of historical records, an expected task value is output on the basis of the prediction model; the expected task value is processed on the basis of a brainstorming optimization algorithm by taking f...

The local community detection is a significant branch of the community detection problems. It aims at finding the local community to which a given starting node belongs. The local community detection plays an important role in analyzing the complex networks and recently has drawn much attention from the researchers. In the past few years, several l...

Combinatorial optimization problems (COPs) are discrete problems arising from aerospace, bioinformatics, manufacturing, and other fields. One of the classic COPs is the scheduling problem. Moreover, these problems are usually multimodal optimization problems with a quantity of global and local optima. As a result, many search algorithms can easily...

The invention discloses a trunking control method, system and UAV for Triassic UAV, which includes collecting designated data, forming a communication network through ZigBee, obtaining the traveling data of a single Triassic UAV, determining an abnormal UAV or a normal UAV according to the traveling data, and the designated data of the abnormal UAV...

Due to rising labor costs, cross-trained worker assignment has become increasingly critical for constructing an efficient and flexible cellular manufacturing systems. Related studies concentrated on assigning skilled workers with different skill levels to tasks according to capacity or cost benefits. However, these studies have yet examined how wor...

As for optimization algorithms, the local optima of multimodal problems always bring difficulties of finding global optimum. This paper proposes a cooperative optimization method that combines differential evolution (DE), univariate sampling, harmony search (HS) and particle swarm optimizer (PSO) to deal with the 100-Digit Challenge of finding glob...

The feature extraction problem for flight data has aroused increasing attention in the practical and the academic aspects. It can reveal the inherent correlation relation among different parameters for the conditional maintenance of the aircraft. However, the high-dimensional and continuous features in the real number field bring challenges to the...

The test task scheduling problem (TTSP) has attracted increasing attention due to the wide range of automatic test systems applications. It is one kind of combinatorial optimization problem with the property of multimodal. There are a lot of global optima and local optima in its huge solution space. Brain storm optimization algorithm (BSO) is a new...

Brain storm optimization (BSO) algorithms is a framework that indicates algorithms using converging operation and diverging operation to locate the optima of optimization problems. Hundreds of articles on the BSO algorithms have been published in different journals and conference proceedings, even though there are more questions than answers. In th...