Partial multi-label learning（PML） allows learning from rich-semantic objects with inaccurate annotations, where a set of candidate labels are assigned to each training example but only some of them are valid. Existi...
详细信息
Partial multi-label learning（PML） allows learning from rich-semantic objects with inaccurate annotations, where a set of candidate labels are assigned to each training example but only some of them are valid. Existing approaches rely on disambiguation to tackle the PML problem, which aims to correct noisy candidate labels by recovering the ground-truth labeling information ahead of prediction model induction. However, this dominant strategy might be suboptimal as it usually needs extra assumptions that cannot be fully satisfied in real-world scenarios. Instead of label correction, we investigate another strategy to tackle the PML problem, where the potential ambiguity in PML data is eliminated by correcting instance features in a label-specific manner. Accordingly, a simple yet effective approach named PASE, i.e., partial multi-label learning via label-specific feature corrections, is proposed. Under a meta-learning framework, PASElearns to exert label-specific feature corrections so that potential ambiguity specific to each class label can be eliminated and the desired prediction model can be induced on these corrected instance features with the provided candidate labels. Comprehensive experiments on a wide range of synthetic and real-world data sets validate the effectiveness of the proposed approach.
This paper tackles uncertainties between planning and actual models. It extends the concept of RCI（robust control invariant） tubes, originally a parameterized representation of closed-loop control robustness in trad...
详细信息
This paper tackles uncertainties between planning and actual models. It extends the concept of RCI（robust control invariant） tubes, originally a parameterized representation of closed-loop control robustness in traditional feedback control, to the domain of motion planning for autonomous vehicles. Thus, closed-loop system uncertainty can be preemptively addressed during vehicle motion planning. This involves selecting collision-free trajectories to minimize the volume of robust invariant tubes. Furthermore, constraints on state and control variables are translated into constraints on the RCI tubes of the closed-loop system, ensuring that motion planning produces a safe and optimal trajectory while maintaining flexibility, rather than solely optimizing for the open-loop nominal model. Additionally, to expedite the solving process, we were inspired by L2gain to parameterize the RCI tubes and developed a parameterized explicit iterative expression for propagating ellipsoidal uncertainty sets within closedloop systems. Furthermore, we applied the pseudospectral orthogonal collocation method to parameterize the optimization problem of transcribing trajectories using high-order Lagrangian polynomials. Finally, under various operating conditions, we incorporate both the kinematic and dynamic models of the vehicle and also conduct simulations and analyses of uncertainties such as heading angle measurement, chassis response, and steering hysteresis. Our proposed robust motion planning framework has been validated to effectively address nearly all bounded uncertainties while anticipating potential tracking errors in control during the planning phase. This ensures fast, closed-loop safety and robustness in vehicle motion planning.
Mixed integer programming is inherently involved in solving a significant number of practical problems. This paper focuses on mixed integer programming, where the objective function is the summation of N functions, an...
详细信息
Mixed integer programming is inherently involved in solving a significant number of practical problems. This paper focuses on mixed integer programming, where the objective function is the summation of N functions, and the constraints include both scalar coupling and set constraints. Given the potentially large scale of these problems, the goal of this work is to propose a distributed method to solve large-scale problems more efficiently. The right-hand side allocation decomposition approach is employed to address the large-scale mixed integer programming problem. Algorithms are then proposed for solving these problems,based on the analysis of the continuity, differentiability, and local convexity properties of the decomposed subproblems. Simulation experiments with randomly generated coefficients demonstrate the superior performance of the proposed algorithms compared to the Gurobi solver, offering higher solution accuracy and faster processing time for large-scale mixed integer programming problems with nonlinear objective and constraint functions.
Current studies against DeepFake attacks are mostly passive methods that detect specific defects of DeepFake algorithms,lacking generalization ***,existing active defense methods only focus on defending against face a...
详细信息
Current studies against DeepFake attacks are mostly passive methods that detect specific defects of DeepFake algorithms,lacking generalization ***,existing active defense methods only focus on defending against face attribute manipulations,and there remain enormous challenges to establishing an active and sustainable defense mechanism for face swap ***,we propose a novel training framework called FSD-GAN(Face Swap Detection based on Generative Adversarial Network),immune to the evolution of face swap ***,FSD-GAN contains three modules:the data processing module,the attack module that generates fake faces only used in training,and the defense module that consists of a fingerprint generator and a fingerprint *** embed the latent noise fingerprints generated by the fingerprint generator into face images,unperceivable to attackers visually and *** an attacker uses these protected faces to perform face swap attacks,these fingerprints will be transferred from training data(protected faces)to generative models(real-world face swap models),and they also exist in generated results(swapped faces).Our discriminator can easily detect latent noise fingerprints embedded in face images,converting the problem of face swap detection to verifying if fingerprints exist in swapped face images or ***,we alternately train the attack and defense modules under an adversarial framework,making the defense module more *** illustrate the effectiveness and robustness of FSD-GAN through extensive experiments,demonstrating that it can confront various face images,mainstream face swap models,and JPEG compression under different qualities.
Data hierarchy,as a hidden property of data structure,exists in a wide range of machine learning applications.A common practice to classify such hierarchical data is first to encode the data in the Euclidean space,and...
详细信息
Data hierarchy,as a hidden property of data structure,exists in a wide range of machine learning applications.A common practice to classify such hierarchical data is first to encode the data in the Euclidean space,and then train a Euclidean ***,such a paradigm leads to a performance drop due to distortion of data embedding in the Euclidean *** relieve this issue,hyperbolic geometry is investigated as an alternative space to encode the hierarchical data for its higher ability to capture the hierarchical *** methods cannot explore the full potential of the hyperbolic geometry,in the sense that such methods define the hyperbolic operations in the tangent plane,causing the distortion of data *** this paper,we develop two novel kernel formulations in the hyperbolic space,with one being positive definite(PD)and another one being indefinite,to solve the classification tasks in hyperbolic *** PD one is defined via mapping the hyperbolic data to the Drury-Arveson(DA)space,which is a special reproducing kernel Hilbert space(RKHS).To further increase the discrimination of the classifier,an indefinite kernel is further defined in the Krein ***,we design a 2-layer nested indefinite kernel which first maps hyperbolic data into the DA spaces,followed by a mapping from the DA spaces to the Krein *** experiments on real-world datasets demonstrate the superiority ofthe proposed kernels.
Electronic auctions(e-auctions)remove the physical limitations of traditional auctions and bring this mechanism to the general ***,most e-auction schemes involve a trusted auctioneer,which is not always credible in **...
详细信息
Electronic auctions(e-auctions)remove the physical limitations of traditional auctions and bring this mechanism to the general ***,most e-auction schemes involve a trusted auctioneer,which is not always credible in *** studies have applied cryptography tools to solve this problem by distributing trust,but they ignore the existence of *** this paper,a blockchain-based Privacy-Preserving and Collusion-Resistant scheme(PPCR)for double auctions is proposed by employing both cryptography and blockchain technology,which is the first decentralized and collusion-resistant double auction scheme that guarantees bidder anonymity and bid privacy.A two-server-based auction framework is designed to support off-chain allocation with privacy preservation and on-chain dispute resolution for collusion resistance.A Dispute Resolution agreement(DR)is provided to the auctioneer to prove that they have conducted the auction correctly and the result is fair and *** addition,a Concise Dispute Resolution protocol(CDR)is designed to handle situations where the number of accused winners is small,significantly reducing the computation cost of dispute *** experimental results confirm that PPCR can indeed achieve efficient collusion resistance and verifiability of auction results with low on-chain and off-chain computational overhead.
Demand response has recently become an essential means for businesses to reduce production costs in industrial ***,the current industrial chain structure has also become increasingly complex,forming new characteristic...
详细信息
Demand response has recently become an essential means for businesses to reduce production costs in industrial ***,the current industrial chain structure has also become increasingly complex,forming new characteristics of multiplex networked industrial *** in real-time electricity prices in demand response propagate through the coupling and cascading relationships within and among these network layers,resulting in negative impacts on the overall energy management ***,existing demand response methods based on reinforcement learning typically focus only on individual agents without considering the influence of dynamic factors on intra and inter-network *** paper proposes a Layered Temporal Spatial Graph Attention(LTSGA)reinforcement learning algorithm suitable for demand response in multiplex networked industrial chains to address this *** algorithm first uses Long Short-Term Memory(LSTM)to learn the dynamic temporal characteristics of electricity prices for ***,LTSGA incorporates a layered spatial graph attention model to evaluate the impact of dynamic factors on the complex multiplex networked industrial chain *** demonstrate that the proposed LTSGA approach effectively characterizes the influence of dynamic factors on intra-and inter-network relationships within the multiplex industrial chain,enhancing convergence speed and algorithm performance compared with existing state-of-the-art algorithms.
Website Fingerprinting(WF)attacks can extract side channel information from encrypted traffic to form a fingerprint that identifies the victim’s destination website,even if traffic is sophisticatedly anonymized by **...
详细信息
Website Fingerprinting(WF)attacks can extract side channel information from encrypted traffic to form a fingerprint that identifies the victim’s destination website,even if traffic is sophisticatedly anonymized by *** offline defenses have been proposed and claimed to have achieved good ***,such work is more of a theoretical optimization study than a technology that can be applied to real-time traffic in the practical *** defenders generate optimized defense schemes only if the complete traffic traces are *** practicality and effectiveness are *** this paper,we provide an in-depth analysis of the difficulties faced in porting existing offline defenses to the online *** then the online WF defense based on the non-targeted adversarial patch is *** reduce the overhead,we use the Gradient-weighted Class Activation Mapping(Grad-CAM)algorithm to identify critical segments that have high contribution to the *** addition,we optimize the adversarial patch generation process by splitting patches and limiting the values,so that the pre-trained patches can be injected and discarded in real-time *** experiments are carried out to evaluate the effectiveness of our *** bandwidth overhead is set to 20%,the accuracies of the two state-of-the-art attacks,DF and Var-CNN,drop to 10.83%and 15.49%,***,we implement the real-time patch traffic injection based on WFPadTools framework in the online scenario,and achieve a defense accuracy of 95.50%with 12.57%time overhead.
Recently,with the increasing complexity of multiplex Unmanned Aerial Vehicles(multi-UAVs)collaboration in dynamic task environments,multi-UAVs systems have shown new characteristics of inter-coupling among multiplex g...
详细信息
Recently,with the increasing complexity of multiplex Unmanned Aerial Vehicles(multi-UAVs)collaboration in dynamic task environments,multi-UAVs systems have shown new characteristics of inter-coupling among multiplex groups and intra-correlation within ***,previous studies often overlooked the structural impact of dynamic risks on agents among multiplex UAV groups,which is a critical issue for modern multi-UAVs communication to *** address this problem,we integrate the influence of dynamic risks on agents among multiplex UAV group structures into a multi-UAVs task migration problem and formulate it as a partially observable Markov *** then propose a Hybrid Attention Multi-agent Reinforcement Learning(HAMRL)algorithm,which uses attention structures to learn the dynamic characteristics of the task environment,and it integrates hybrid attention mechanisms to establish efficient intra-and inter-group communication aggregation for information extraction and group *** results show that in this comprehensive and challenging model,our algorithm significantly outperforms state-of-the-art algorithms in terms of convergence speed and algorithm performance due to the rational design of communication mechanisms.
Label distribution learning(LDL) has shown advantages over traditional single-label learning(SLL) in many realworld applications, but its superiority has not been theoretically understood. In this paper, we attempt to...
详细信息
Label distribution learning(LDL) has shown advantages over traditional single-label learning(SLL) in many realworld applications, but its superiority has not been theoretically understood. In this paper, we attempt to explain why LDL generalizes better than SLL. Label distribution has rich supervision information such that an LDL method can still choose the sub-optimal label from label distribution even if it neglects the optimal one. In comparison, an SLL method has no information to choose from when it fails to predict the optimal label. The better generalization of LDL can be credited to the rich information of label distribution. We further establish the label distribution margin theory to prove this explanation; inspired by the theory,we put forward a novel LDL approach called LDL-LDML. In the experiments, the LDL baselines outperform the SLL ones, and LDL-LDML achieves competitive performance against existing LDL methods, which support our explanation and theories in this paper.
暂无评论