The creator economy has revolutionized the way individuals can profit through online platforms. In this paper, we initiate the study of online learning in the creator economy by modeling the creator economy as a three-party game between the users, platform, and content creators, with the platform interacting with the content creator under a principal-agent model through contracts to encourage better content. Additionally, the platform interacts with the users to recommend new content, receive an evaluation, and ultimately profit from the content, which can be modeled as a recommender system. Our study aims to explore how the platform can jointly optimize the contract and recommender system to maximize the utility in an online learning fashion. We primarily analyze and compare two families of contracts: return-based contracts and feature-based contracts. Return-based contracts pay the content creator a fraction of the reward the platform gains. In contrast, feature-based contracts pay the content creator based on the quality or features of the content, regardless of the reward the platform receives. We show that under smoothness assumptions, the joint optimization of return-based contracts and recommendation policy provides a regret $\Theta(T^{2/3})$. For the feature-based contract, we introduce a definition of intrinsic dimension $d$ to characterize the hardness of learning the contract and provide an upper bound on the regret $\mathcal{O}(T^{(d+1)/(d+2)})$. The upper bound is tight for the linear family.
As financial markets grow increasingly complex and volatile, time-series-based stock price forecasting has become a critical research focus in the field of finance. Traditional forecasting methods face significant limitations in handling nonlinear and high-dimensional data, while neural networks (NNs) have demonstrated great potential due to their powerful feature extraction and pattern recognition capabilities. Although several existing surveys discuss the applications of NNs in stock forecasting, they often lack a detailed examination of models that use time-series data as input and fail to cover the latest research developments. In response, this paper reviews relevant literature from 2015 to 2025 and classifies time-series-based stock forecasting methods into four categories: NNs, recurrent NNs (RNNs), convolutional NNs (CNNs), Transformers and other models. We analyze their performance under different market conditions, highlight strengths and limitations, and identify recent trends in model design. Our findings show that hybrid architectures and attention-based models consistently achieve superior forecasting stability and adaptability across volatile market scenarios. This survey offers a systematic reference for researchers and practitioners and outlines promising future research directions.
In dynamic scenes, the pose estimation and map consistency of visual simultaneous localisation and mapping (visual SLAM) are affected by intermittent changes in object motion states. An adaptive motion-state estimation and feature-reuse mechanism is proposed which restores features once objects become stationary. Camera ego-motion is compensated via projection-based point-to-point red-green-blue-depth (RGB-D) Iterative Closest Point; the alignment residual yields a short-term jitter score. An Extended Kalman Filter fuses the centre-pixel trajectory and depth of the object, using depth innovation as strong evidence to suppress false triggers. Applied adaptive decision thresholds involve resolution, ego-motion intensity, jitter, and reference depth, and are combined with dual/single triggering and hysteresis to achieve robust switching. When an object is considered static, its feature points are reused. On the Bonn RGB-D Dynamic Dataset (BONN) and TUM RGB-D SLAM Dataset and Benchmark (TUM), the proposed method matches or exceeds baselines: In intermittent-motion-dominated BONN sequences Placing_non_box, it reduces the root-mean-square of the absolute trajectory error (ATE-RMSE) by 27% relative to the baseline, remains comparable to Ellipsoid-SLAM on TUM, and consistently outperforms ORB-SLAM3 in dynamic scenes. The hysteresis counter reading on Placing_non_box2 shows that the proposed method can reduce the motion-state misclassification rate by nearly 40%. From the ablation experiment results, we confirm that adaptive thresholds yield the most significant optimisation effect. The approach improves robustness and map completeness in dynamic environments without degrading performance in low-dynamic settings.
The balancing market in the energy sector plays a critical role in physically and financially balancing the supply and demand. Modeling dynamics in the balancing market can provide valuable insights and prognosis for power grid stability and secure energy supply. While complex machine learning models can achieve high accuracy, their “black-box” nature severely limits the model interpretability. In this paper, we explore the trade-off between model accuracy and interpretability for the energy balancing market. Particularly, we take the example of forecasting manual frequency restoration reserve (mFRR) activation price in the balancing market using real market data from different energy price zones. We explore the interpretability of mFRR forecasting using two models: extreme gradient boosting (XGBoost) machine and explainable boosting machine (EBM). We also integrate the two models, and we benchmark all the models against a baseline naive model. Our results show that EBM provides forecasting accuracy comparable to XGBoost while yielding a considerable level of interpretability. Our analysis also underscores the challenge of accurately predicting the mFRR price for the instances when the activation price deviates significantly from the spot price. Importantly, EBM's interpretability features reveal insights into non-linear mFRR price drivers and regional market dynamics. Our study demonstrates that EBM is a viable and valuable interpretable alternative to complex black-box AI models in the forecast for the balancing market.
This paper presents a comprehensive overview of distributed Nash equilibrium (NE) seeking algorithms in non-cooperative games for multi-agent systems (MASs), with a distinct emphasis on the dynamic control perspective. It specifically focuses on the research addressing distributed NE seeking problems in which agents are governed by heterogeneous dynamics. The paper begins by introducing fundamental concepts of general non-cooperative games and the NE, along with definitions of specific game structures such as aggregative games and multi-cluster games. It then systematically reviews existing studies on distributed NE seeking for various classes of MASs from the viewpoint of agent dynamics, including first-order, second-order, high-order, linear, and Euler-Lagrange (EL) systems. Furthermore, the paper highlights practical applications of these theoretical advances in cooperative control scenarios involving autonomous systems with complex dynamics, such as autonomous surface vessels, autonomous aerial vehicles, and other autonomous vehicles. Finally, the paper outlines several promising directions for future research.
Optimal impulse control and impulse games provide the cutting-edge frameworks for modeling systems where control actions occur at discrete time points,and optimizing objectives under discontinuous interventions.This review synthesizes the theoretical advancements,computational approaches,emerging challenges,and possible research directions in the field.Firstly,we briefly review the fundamental theory of continuous-time optimal control,including Pontryagin's maximum principle(PMP)and dynamic programming principle(DPP).Secondly,we present the foun-dational results in optimal impulse control,including necessary conditions and sufficient conditions.Thirdly,we systematize impulse game methodo-logies,from Nash equilibrium existence theory to the connection between Nash equilibrium and systems stability.Fourthly,we summarize the nu-merical algorithms including the intelligent computation approaches.Finally,we examine the new trends and challenges in theory and appli-cations as well as computational considerations.
Spiking neural networks(SNN)represent a paradigm shift toward discrete,event-driven neural computation that mirrors biological brain mechanisms.This survey systematically examines current SNN research,focusing on training methodologies,hardware implementations,and practical applications.We analyze four major training paradigms:ANN-to-SNN conversion,direct gradient-based training,spike-timing-dependent plasticity(STDP),and hybrid approaches.Our review encompasses major specialized hardware platforms:Intel Loihi,IBM TrueNorth,SpiNNaker,and BrainScaleS,analyzing their capabilities and constraints.We survey applications spanning computer vision,robotics,edge computing,and brain-computer interfaces,identifying where SNN provide compelling advantages.Our comparative analysis reveals SNN offer significant energy efficiency improvements(1000-10000×reduction)and natural temporal processing,while facing challenges in scalability and training complexity.We identify critical research directions including improved gradient estimation,standardized benchmarking protocols,and hardware-software co-design approaches.This survey provides researchers and practitioners with a comprehensive understanding of current SNN capabilities,limitations,and future prospects.
IntuiGrasp is a novel three-fingered dexterous hand that pioneers bio-inspired demonstrations with intuitive priors (BDIP) to bridge the gap between human tactile intuition and robotic execution. Unlike conventional programming, BDIP leverages human's innate priors (e.g., “A pack of tissues requires gentle grasps, cups demand firm contact”) by enabling real-time transfer of gesture and force policies during physical demonstration. When a human demonstrator wears IntuiGrasp, driven rings provide real-time haptic feedback on contact stress and slip, while integrated tactile sensors translate these human policies into image data, offering valuable data for imitation learning. In this study, human teachers use IntuiGrasp to demonstrate how to grasp three types of objects: a cup, a crumpled tissue pack, and a thin playing card. IntuiGrasp translates the policies for grasping these objects into image information that describes tactile sensations in real time.
Formalizing complex processes and phenomena of a real-world problem may require a large number of variables and constraints, resulting in what is termed a large-scale optimization problem. Nowadays, such large-scale optimization problems are solved using computing machines, leading to an enormous computational time being required, which may delay deriving timely solutions. Decomposition methods, which partition a large-scale optimization problem into lower-dimensional subproblems, represent a key approach to addressing time-efficiency issues. There has been significant progress in both applied mathematics and emerging artificial intelligence approaches on this front. This work aims at providing an overview of the decomposition methods from both the mathematics and computer science points of view. We also remark on the state-of-the-art developments and recent applications of the decomposition methods, and discuss the future research and development perspectives.
This paper considers the swarm vigilance problem for multi-agent systems (MAS), where multiple agents are deployed within a rectangular region for perception-based vigilance. There are two main challenges, namely the task allocation for vigilance roles and the coverage planning of the perception ranges. Firstly, vigilance behavioral patterns and processes in animal populations within natural habitats are investigated. Inspired by these biological vigilance behaviors, an efficient vigilance task allocation model for MAS is proposed. Secondly, the subsequent optimization of task layouts can achieve efficient surveillance coverage with fewer agents, minimizing resource consumption. Thirdly, an improved particle swarm optimization (IPSO) algorithm is proposed, which incorporates fitness-driven adaptive inertia weight dynamics. According to simulation analysis and comparative studies, optimal parameter configurations for Genetic Algorithm (GA) and IPSO are determined. Finally, the results indicate the proposed IPSO's superior performance to both GA and standard particle swarm optimization (PSO) in vigilance task allocation optimization, with satisfying advantages in computational efficiency and solution quality.
Inverse reinforcement learning optimal control is under the framework of learner-expert. The learner system can imitate the expert system's demonstrated behaviors and does not require the predefined cost function, so it can handle optimal control problems effectively. This paper proposes an inverse reinforcement learning optimal control method for Takagi-Sugeno (T-S) fuzzy systems. Based on learner systems, an expert system is constructed, where the learner system only knows the expert system's optimal control policy. To reconstruct the unknown cost function, we firstly develop a model-based inverse reinforcement learning algorithm for the case that systems dynamics are known. The developed model-based learning algorithm is consists of two learning stages: an inner reinforcement learning loop and an outer inverse optimal control loop. The inner loop desires to obtain optimal control policy via learner's cost function and the outer loop aims to update learner's state-penalty matrices via only using expert's optimal control policy. Then, to eliminate the requirement that the system dynamics must be known, a data-driven integral learning algorithm is presented. It is proved that the presented two algorithms are convergent and the developed inverse reinforcement learning optimal control scheme can ensure the controlled fuzzy learner systems to be asymptotically stable. Finally, we apply the proposed fuzzy optimal control to the truck-trailer system, and the computer simulation results verify the effectiveness of the presented approach.
The application of visual-language large models in the field of medical health has gradually become a research focus. The models combine the capability for image understanding and natural language processing, and can simultaneously process multi-modality data such as medical images and medical reports. These models can not only recognize images, but also understand the semantic relationship between images and texts, effectively realize the integration of medical information, and provide strong support for clinical decision-making and disease diagnosis. The visual-language large model has good performance for specific medical tasks, and also shows strong potential and high intelligence in the general task models. This paper provides a comprehensive review of the visual-language large model in the field of medical health. Specifically, this paper first introduces the basic theoretical basis and technical principles. Then, this paper introduces the specific application scenarios in the field of medical health, including modality fusion, semi-supervised learning, weakly supervised learning, unsupervised learning, cross-domain model and general models. Finally, the challenges including insufficient data, interpretability, and practical deployment are discussed. According to the existing challenges, four potential future development directions are given.
The event-triggered mechanism serves as an effective discontinuous control strategy for addressing the consensus tracking problem in multi-agent systems (MASs). This approach optimizes energy consumption by updating the controller only when some observed errors exceed a predefined threshold. Considering the influence of noise on agent dynamics in complex control environments, this study investigates an event-triggered control scheme for stochastic MASs, where noise is modeled as Brownian motion. Furthermore, the communication topology of the stochastic MASs is assumed to exhibit a Markovian switching mechanism. Analytical criteria are derived to guarantee consensus tracking in the mean square sense, and a numerical example is provided to validate the effectiveness of the proposed control methods.
This paper addresses the consensus problem of nonlinear multi-agent systems subject to external disturbances and uncertainties under denial-of-service (DoS) attacks. Firstly, an observer-based state feedback control method is employed to achieve secure control by estimating the system's state in real time. Secondly, by combining a memory-based adaptive event-triggered mechanism with neural networks, the paper aims to approximate the nonlinear terms in the networked system and efficiently conserve system resources. Finally, based on a two-degree-of-freedom model of a vehicle affected by crosswinds, this paper constructs a multi-unmanned ground vehicle (Multi-UGV) system to validate the effectiveness of the proposed method. Simulation results show that the proposed control strategy can effectively handle external disturbances such as crosswinds in practical applications, ensuring the stability and reliable operation of the Multi-UGV system.
Cooperative multi-agent reinforcement learning (MARL) is a key technology for enabling cooperation in complex multi-agent systems. It has achieved remarkable progress in areas such as gaming, autonomous driving, and multi-robot control. Empowering cooperative MARL with multi-task decision-making capabilities is expected to further broaden its application scope. In multi-task scenarios, cooperative MARL algorithms need to address 3 types of multi-task problems: reward-related multi-task, arising from different reward functions; multi-domain multi-task, caused by differences in state and action spaces, state transition functions; and scalability-related multi-task, resulting from the dynamic variation in the number of agents. Most existing studies focus on scalability-related multi-task problems. However, with the increasing integration between large language models (LLMs) and multi-agent systems, a growing number of LLM-based multi-agent systems have emerged, enabling more complex multi-task cooperation. This paper provides a comprehensive review of the latest advances in this field. By combining multi-task reinforcement learning with cooperative MARL, we categorize and analyze the 3 major types of multi-task problems under multi-agent settings, offering more fine-grained classifications and summarizing key insights for each. In addition, we summarize commonly used benchmarks and discuss future directions of research in this area, which hold promise for further enhancing the multi-task cooperation capabilities of multi-agent systems and expanding their practical applications in the real world.
This paper explores the issue of secure synchronization control in piecewise-homogeneous Markovian jump delay neural networks affected by denial-of-service (DoS) attacks. Initially, a novel memory-based adaptive event-triggered mechanism (MBAETM) is designed based on sequential growth rates, focusing on event-triggered conditions and thresholds. Subsequently, from the perspective of defenders, non-periodic DoS attacks are re-characterized, and a model of irregular DoS attacks with cyclic fluctuations within time series is further introduced to enhance the system's defense capabilities more effectively. Additionally, considering the unified demands of network security and communication efficiency, a resilient memory-based adaptive event-triggered mechanism (RMBAETM) is proposed. A unified Lyapunov-Krasovskii functional is then constructed, incorporating a loop functional to thoroughly consider information at trigger moments. The master-slave system achieves synchronization through the application of linear matrix inequality techniques. Finally, the proposed methods' effectiveness and superiority are confirmed through four numerical simulation examples.
For large-scale heterogeneous multi-agent systems (MASs) with characteristics of dense-sparse mixed distribution, this paper investigates the practical finite-time deployment problem by establishing a novel cross-species bionic analytical framework based on the partial differential equation-ordinary differential equation (PDE-ODE) approach. Specifically, by designing a specialized network communication protocol and employing the spatial continuum method for densely distributed agents, this paper models the tracking errors of densely distributed agents as a PDE equivalent to a human disease transmission model, and that of sparsely distributed agents as several ODEs equivalent to the predator population models. The coupling relationship between the PDE and ODE models is established through boundary conditions of the PDE, thereby forming a PDE-ODE-based tracking error model for the considered MASs. Furthermore, by integrating adaptive neural control scheme with the aforementioned biological models, a “Flexible Neural Network” endowed with adaptive and self-stabilized capabilities is constructed, which acts upon the considered MASs, enabling their practical finite-time deployment. Finally, effectiveness of the developed approach is illustrated through a numerical example.
The rapid advancement of deep learning and the emergence of large-scale neural models, such as bidirectional encoder representations from transformers (BERT), generative pre-trained transformer (GPT), and large language model Meta AI (LLaMa), have brought significant computational and energy challenges. Neuromorphic computing presents a biologically inspired approach to addressing these issues, leveraging event-driven processing and in-memory computation for enhanced energy efficiency. This survey explores the intersection of neuromorphic computing and large-scale deep learning models, focusing on neuromorphic models, learning methods, and hardware. We highlight transferable techniques from deep learning to neuromorphic computing and examine the memory-related scalability limitations of current neuromorphic systems. Furthermore, we identify potential directions to enable neuromorphic systems to meet the growing demands of modern AI workloads.
This article briefly reviews the topic of complex network synchronization, with its graph-theoretic criterion, showing that the homogeneous and symmetrical network structures are essential for optimal synchronization. Furthermore, it briefly reviews the notion of higher-order network topologies and shows their promising potential in application to evaluating the optimality of network synchronizability.
Cyber-physical systems (CPSs) are regarded as the backbone of the fourth industrial revolution, in which communication, physical processes, and computer technology are integrated. In modern industrial systems, CPSs are widely utilized across various domains, such as smart grids, smart healthcare systems, smart vehicles, and smart manufacturing, among others. Due to their unique spatial distribution, CPSs are highly vulnerable to cyber-attacks, which may result in severe performance degradation and even system instability. Consequently, the security concerns of CPSs have attracted significant attention in recent years. In this paper, a comprehensive survey on the security issues of CPSs under cyber-attacks is provided. Firstly, mathematical descriptions of various types of cyber-attacks are introduced in detail. Secondly, two types of secure estimation and control processing schemes, including robust methods and active methods, are reviewed. Thirdly, research findings related to secure control and estimation problems for different types of CPSs are summarized. Finally, the survey is concluded by outlining the challenges and suggesting potential research directions for the future.
