Traditional multiobjective evolutionary algorithms face a great challenge when dealing with many objectives. This is due to high proportion of nondominated solutions in the population and low selection pressure toward Pareto front. In order tackle this issue, series indicator-based have been proposed guide search process However, single indicator might be biased lead converge subregion paper, multi-indicator-based algorithm for many-objective optimization problems. The algorithm, namely...

Multi-Objective Evolutionary Algorithms have been deeply studied in the research community and widely used real-world applications. However, performance of traditional Pareto-based MOEAs, such as NSGA-II SPEA2, may deteriorate when tackling Many-Objective Problems, which refer to problems with at least four objectives. The main cause for degradation lies that high-proportional non-dominated solutions severely weaken differentiation ability Pareto-dominance. This lead stagnation. Two Archive...

Multi-objective optimization problems (MOPs) containing a large number of decision variables, which are also known as large-scale multi-objective (LSMOPs), pose great challenges to most existing evolutionary algorithms. This is mainly because that high dimensional space degrades the effectiveness search operators notably, and balancing convergence diversity becomes challenging task. In this paper, we propose two-population based algorithm for named LSTPA. proposed algorithm, solutions...

Cognitive diagnosis models (CDMs) are designed to learn students' mastery levels using their response logs. CDMs play a fundamental role in online education systems since they significantly influence downstream applications such as teachers' guidance and computerized adaptive testing. Despite the success achieved by existing CDMs, we find that suffer from thorny issue learned too similar. This issue, which refer oversmoothing, could diminish CDMs' effectiveness tasks. comprise two core...

Surrogate-assisted Evolutionary Algorithm (SAEA) is an essential method for solving expensive problems. Utilizing surrogate models to substitute the optimization function can significantly reduce reliance on evaluations during search process, thereby lowering costs. The construction of a critical component in SAEAs, with numerous machine learning algorithms playing pivotal role model-building phase. This paper introduces Kolmogorov-Arnold Networks (KANs) as within examining their application...

A series of surrogate-assisted evolutionary algorithms (SAEAs) have been proposed for expensive multi-objective optimization problems (EMOPs), building cheap surrogate models to replace the real function evaluations. However, search efficiency these SAEAs is not yet satisfactory. More efforts are needed further exploit useful information from evaluations in order better guide process. Facing this challenge, paper proposes a Hyperbolic Neural Network (HNN) based preselection operator...

With the explosively increase of information and products, recommender systems have played a more important role in recent years. Various recommendation algorithms, such as content-based methods collaborative filtering methods, been proposed. There are number performance metrics for evaluating systems, considering only precision or diversity might be inappropriate. However, to best our knowledge, no existing work has considered with many objectives. In this paper, we model many-objective...

Due to computationally and/or financially costly evaluation, tackling expensive multi-objective optimization problems is quite challenging for evolutionary algorithms. One popular approach these building cheap surrogate models replace the real function evaluations. To this end, various kinds of surrogate-assisted algorithms (SAEAs) have been proposed, which predict fitness values, classifications, or relation candidate solutions. However, off-spring generation, despite its important role in...

A series of surrogate-assisted evolutionary algorithms (SAEAs) have been proposed to handle expensive multi-objective optimization problems (EMOPs). However, the surrogate these SAEAs is underutilized a large extent, which limits search efficiency algorithms. To be specific, existing do not sufficiently exploit estimated solution quality information from models during offspring generation. address this issue, paper proposes an SAEA framework named EXO-SAEA (EXplanation Operator based...

Multi-objective combinatorial optimization (MOCO) problems are prevalent in various real-world applications. Most existing neural methods for MOCO rely solely on decomposition and utilize precise hypervolume to enhance diversity. However, these often approximate only limited regions of the Pareto front spend excessive time diversity enhancement because ambiguous time-consuming calculation. To address limitations, we design a Geometry-Aware set Learning algorithm named GAPL, which provides...

Pointer Network (PtrNet) is a specific neural network for solving Combinatorial Optimization Problems (COPs). While PtrNets offer real-time feed-forward inference complex COPs instances, its quality of the results tends to be less satisfactory. One possible reason that such issue suffers from lack global search ability gradient descent, which frequently employed in traditional PtrNet training methods including both supervised learning and reinforcement learning. To improve performance...