Update Rule Research Articles

R-DOCO: Resilient Distributed Online Convex Optimization Against Adversarial Attacks

This paper addresses the problem of distributed constrained optimization in a multi-agent system where some agents may deviate from the prescribed update rules due to failures or malicious adversarial attacks. The objective is to minimize the collective cost of the unattacked agents while respecting the constraint limitations. To tackle this, we propose a resilient distributed projected gradient descent algorithm for online optimization that achieves sublinear individual regret, defined as the difference between the online and offline solutions. Additionally, we extend the cost function from convex combinations to more general distributed optimization scenarios. The proposed algorithm demonstrates resilience under adversarial conditions, allowing it to handle an unknown number of adversarial nodes while maintaining performance. Compared to existing methods, this approach offers a robust solution to adversarial attacks in constrained distributed optimization problems.

Evolution of cooperation in the public goods game with Q-learning

Recent paradigm shifts from imitation learning to reinforcement learning (RL) is shown to be productive in understanding human behaviors. In the RL paradigm, individuals search for optimal strategies through interaction with the environment to make decisions. This implies that gathering, processing, and utilizing information from their surroundings is crucial. However, existing studies on public good games using the Q-learning algorithm typically adopt a self-regarding setup, where individuals adjust their strategies based solely on their own strategic information, neglecting the environmental factors. In this work, we investigate the evolution of cooperation with the multiplayer game — the public goods game using the Q-learning algorithm by leveraging the environmental information. Specifically, the decision-making of players is based on the cooperation information in their neighborhood. Our results show that cooperation is more likely to emerge compared to the case of imitation learning by using Fermi-function-based update rule. Of particular interest is the observation of an anomalous non-monotonic dependence which is revealed when voluntary participation is further introduced. The analysis of the Q-table explains the mechanisms behind the cooperation evolution. Our findings indicate the fundamental role of environment information in the RL paradigm to understand the evolution of cooperation, and human behaviors in general.

Asymmetric iterated prisoner’s dilemma on weighted complex networks and evolutionary strategies analysis

Abstract In real-world scenarios, individuals often cooperate for mutual benefit. However, differences in wealth, reputation, and rationality can lead to varying outcomes for similar actions. Besides, in complex social networks, an individual’s choices are frequently influenced by their neighbors. To explore the evolution of strategies in realistic settings, we conduct repeated asymmetric iterated prisoner’s dilemma experiments on weighted networks using a Memory-one strategy framework and different strategy update rules. During the strategy evolution on the network, two key strategies emerge, and we name them as ‘self-bad, partner-worse’ and the ‘altruists’. Then, we perform separate evolutionary experiments on several strong strategies on corresponding networks and find that strategy ‘self-bad, partner-worse’ can still stand out from the dominant strategies. Finally, by introducing optimization mechanisms, we increase the cooperation levels among individuals within the group. The models utilize in these studies diverge from conventional approaches, scrutinizing the evolutionary process at a macroscopic scale. These findings broaden the scope of evolutionary games and furnish a foundation for addressing real-world challenges.

Distributed adaptive moving horizon estimation for multi-sensor networks subject to quantization effects

This paper investigates the distributed state estimation problem for multi-sensor networks with quantized measurements. Within the Bayesian framework, a distributed adaptive moving horizon estimation algorithm is developed. Unlike the existing methods regarding quantized errors roughly as bounded uncertainties, the posterior distributions of the errors are demanded to be derived. To overcome the difficulty of evaluating the posterior distributions for series of the states and quantized errors jointly, the variational Bayesian methodology is adopted to approximate the true distributions. Based on the fixed-point iteration method, the update rules are analytically derived, with the convergence criterion provided. Furthermore, by incorporating the average consensus algorithm into the prediction process, all sensors can achieve consensus on their estimates in a distributed manner. Finally, a numerical example of target tracking under logarithmic and uniform quantization effects is given to illustrate the validity of the proposed algorithm.

Lessons for expert psychiatric witnesses from recent judgments and updated procedural rules

SUMMARY The past 5 years have seen numerous court judgments and changes to rules and procedures that relate to the work of expert psychiatric witnesses in the British Isles. This article outlines these changes, pointing out their implications for the expert witness, and highlights pertinent judgments in over 100 court and tribunal cases.

Batch channel normalized-CWGAN with Swin Transformer for imbalanced data fault diagnosis of rotating machinery

Abstract In real scenarios, rotating machinery is mainly operated in optimal condition, leading to fault data scarce and difficult to collect. This issue results in imbalanced data, significantly limiting the effectiveness of intelligent fault diagnosis methods. To address this issue, a novel fault diagnosis method for rotating machinery is proposed in this paper, which combines the batch channel normalized conditional wasserstein generative adversarial network (BCN-CWGAN) with Swin Transformer. Firstly, the one-dimensional vibration signal is preprocessed into two-dimensional feature images using a symmetrized dot pattern (SDP). Subsequently, self-attention mechanism and deep feature learning module constructed by DenseNet are integrated into the generator of GAN to acquire more discriminative feature information. Meanwhile, the discriminator of GAN is combined with batch channel normalization strategy, which further enhances the generalization ability. Besides, a two time-scale update rule strategy enhances training stability and convergence speed by updating model parameters at different time scales. Then, the data augmentation capability of BCN-CWGAN is used to generate high-quality fault samples to augment the imbalanced dataset. Finally, Swin Transformer is combined to achieve accurate fault diagnosis. The performance enhancement of the proposed method is verified through comparison and diagnosis results of two engineering experiments, demonstrating its substantial value for research in engineering practice. With the proposed data augmentation method, the average accuracy of A1, B1, C1, and D1 datasets in experiment 1 reached 99.24%, 98.85%, 96.78%, and 96.04%, respectively. Meanwhile, the proposed method achieved the best accuracy in experiment 2.

Survival and Evolutionary Adaptation of Populations Under Disruptive Habitat Change: A Study With Darwinian Cellular Automata.

The evolution of living beings with continuous and consistent progress toward adaptation and ways to model evolution along principles as close as possible to Darwin's are important areas of focus in Artificial Life. Though genetic algorithms and evolutionary strategies are good methods for modeling selection, crossover, and mutation, biological systems are undeniably spatially distributed processes in which living organisms interact with locally available individuals rather than with the entire population at once. This work presents a model for the survival of organisms during a change in the environment to a less favorable one, putting them at risk of extinction, such as many organisms experience today under climate change or local habitat loss or fragmentation. Local spatial structure of resources and environmental quality also impacts the capacity of an evolving population to adapt. The problem is considered on a probabilistic cellular automaton with update rules based on the principles of genetic algorithms. To carry out simulations according to the described model, the Darwinian cellular automata are introduced, and the software has been designed with the code available open source. An experimental evaluation of the behavioral characteristics of the model was carried out, completed by a critical evaluation of the results obtained, parametrically describing conditions and thresholds under which extinction or survival of the population may occur.

Equilibrium seeking of higher-order networks under facet cover constraints.

Agent intelligence involves specific requirements for social attributes. Intelligent agents make their decisions based on the groups they are part of, tend to satisfy co-members, and enlarge their own benefits. A fundamental question is whether this form of subgroup decision-making accommodate each individual's preferences. In this paper, we examine the evolution of an anticoordination game on a higher-order network in the form of a simplicial complex in relation to the facet cover problem, which ensures that each subgroup yields a positive benefit. We introduce and apply the facet update rule to regulate nodes' group-based interactions. We identify the payoff parameter condition that a strict Nash equilibrium (SNE) satisfies a facet cover. The proposed facet update rule enables the activated facet to reach a facet equilibrium, and all nodes would converge to an SNE with no more than 2m strategy switches, where m is the number of nodes in the simplicial complex. Additionally, we analyze the convergence of the asynchronous update rule, which can be seen as a special case of the facet update rule. Our simulations and extensive examples reveal that the SNE achieved by the facet update rule, on average, covers fewer nodes compared to the asynchronous update rule.

Reconstructing redshift distributions with photometric galaxy clustering

The accurate determination of the true redshift distributions in tomographic bins is critical for cosmological constraints from photometric surveys. The proposed redshift self-calibration method, which utilizes the photometric galaxy clustering alone, is highly convenient and avoids the challenges from incomplete or unrepresentative spectroscopic samples in external calibration. However, the imperfection of the theoretical approximation on broad bins as well as the flaw of the algorithm in previous work [1] risk the accuracy and application of the method. In this paper, we propose the improved self-calibration algorithm that incorporates novel update rules, which effectively accounts for heteroskedastic weights and noisy data with negative values. The improved algorithm greatly expands the application range of self-calibration method and accurately reconstructs the redshift distributions for various mock data. Using the luminous red galaxy (LRG) sample of the Dark Energy Spectroscopic Instrument (DESI) survey, we find that the reconstructed results are comparable to the state-of-the-art external calibration. This suggests the exciting prospect of using photometric galaxy clustering to reconstruct redshift distributions in the cosmological analysis of survey data.

Stability of networked control systems under malicious attacks and bit rate constraints

This article investigates the stability of networked control systems under two types of malicious attacks and bit rate constraints. The malicious attacks may lead to parameter mismatches, including scaling variable mismatches between the encoder and decoder sides and discrepancies between the data sent by the encoder and received by the decoder, causing system instability. To address these issues, scaling variable update rules are proposed for the encoder and decoder sides separately to mitigate the adverse effects of malicious attacks. A control strategy featuring a compensation term is designed to counteract the detrimental impacts of the mismatch. A lower bound on the bit rate is established to prevent quantization saturation and achieve exponential stability of the system. Finally, confirmatory and comparative simulations showcase the effectiveness of the theoretical insights and the merits of the proposed strategies, respectively.

Requester mobility for mobile crowdsensing system: A dynamic alliance-based incentive mechanism

In the Mobile Crowdsensing (MCS) system, due to the heterogeneity of requesters, they are in mobile which means requesters will join or leave the MCS system at different times and their data demands are time-varying. The uncertainty of requesters caused by requester mobility will generate the instability of the match between requesters and participants which means the established matching algorithm cannot be completed due to the dissatisfaction of requesters and participants caused by the changes in the intensity of competition. For requester mobility, we design a dynamic alliance-based incentive mechanism where requesters can leave, join the MCS system separately, and change their data needs. To the instability, we divide the dynamic mechanism into different stages and will update the matching rules in each stage. A unique algorithm based on the dynamic Stackelberg game and the corresponding updated algorithm is proposed to analyze the matching strategies of requesters and participants to make an optimal match. By proving the stability of the updated rules, we guarantee the stability of the match with requester mobility. Through numerical analysis, alliance formation can significantly reduce weak requesters’ costs by at most 90%. Besides, in our mechanism any requester participates in the game at most twice, the chosen rate can be up to 100%.

Noise-resistant sharpness-aware minimization in deep learning

Sharpness-aware minimization (SAM) aims to enhance model generalization by minimizing the sharpness of the loss function landscape, leading to a robust model performance. To protect sensitive information and enhance privacy, prevailing approaches add noise to models. However, additive noises would inevitably degrade the generalization and robustness of the model. In this paper, we propose a noise-resistant SAM method, based on a noise-resistant parameter update rule. We analyze the convergence and noise resistance properties of the proposed method under noisy conditions. We elaborate on experimental results with several networks on various benchmark datasets to demonstrate the advantages of the proposed method with respect to model generalization and privacy protection.

XGrad: Boosting Gradient-Based Optimizers With Weight Prediction.

In this paper, we propose a general deep learning training framework XGrad which introduces weight prediction into the popular gradient-based optimizers to boost their convergence and generalization when training the deep neural network (DNN) models. In particular, ahead of each mini-batch training, the future weights are predicted according to the update rule of the used optimizer and are then applied to both the forward pass and backward propagation. In this way, during the whole training period, the optimizer always utilizes the gradients w.r.t. the future weights to update the DNN parameters, making the gradient-based optimizer achieve better convergence and generalization compared to the original optimizer without weight prediction. XGrad is rather straightforward to implement yet pretty effective in boosting the convergence of gradient-based optimizers and the accuracy of DNN models. Empirical results concerning five popular optimizers including SGD with momentum, Adam, AdamW, AdaBelief, and AdaM3 demonstrate the effectiveness of our proposal. The experimental results validate that XGrad can attain higher model accuracy than the baseline optimizers when training the DNN models.

AOCBLS: A novel active and online learning system for ECG arrhythmia classification with less labeled samples

Electrocardiogram (ECG) is a pivotal determinant of cardiac arrhythmia. In practice, ECG data are often acquired as continuous unlabeled chunks with high labeling costs and severe class imbalances. This poses formidable challenges to existing offline learning methods. To address these issues, we propose a novel model called active online class-specific broad learning system (AOCBLS) to learn streaming imbalanced ECG and classify arrhythmias. First, a two-stage active learning strategy combining diversity and uncertainty queries was designed to select the most valuable beats from the current chunk for further labeling and learning, thereby reducing labeling costs. Subsequently, a stable training weight update rule was developed to address the dynamic imbalances in the streaming data. Finally, an incremental learning algorithm was proposed to update network weights online using selected beats, eliminating the need for retraining from scratch. Thus, the AOCBLS can effectively learn online from imbalanced data streams with less label information. Intra-patient experiments on the benchmark MIT-BIH arrhythmia database showed our model achieved an overall accuracy of 99.10% and a G-mean of 96.35% in identifying the four AAMI-recommended beat types, with only 10% additional annotation, outperforming advanced methods at lower labeling cost. Additionally, patient-specific experimental results on the benchmark database and two other databases, INCARTDB and SVDB, suggest the potential of AOCBLS for personalized ECG monitoring on portable devices.

Lossy Image Compression with Stochastic Quantization

Introduction. Lossy image compression algorithms play a crucial role in various domains, including graphics, and image processing. As image information density increases, so do the resources required for processing and transmission. One of the most prominent approaches to address this challenge is color quantization, proposed by Orchard et al. (1991). This technique optimally maps each pixel of an image to a color from a limited palette, maintaining image resolution while significantly reducing information content. Color quantization can be interpreted as a clustering problem (Krishna et al. (1997), Wan (2019)), where image pixels are represented in a three-dimensional space, with each axis corresponding to the intensity of an RGB channel. The purpose of the paper. Scaling of traditional algorithms like K-Means can be challenging for large data, such as modern images with millions of colors. This paper reframes color quantization as a three-dimensional stochastic transportation problem between the set of image pixels and an optimal color palette, where the number of colors is a predefined hyperparameter. We employ Stochastic Quantization (SQ) with a seeding technique proposed by Arthur et al. (2007) to enhance the scalability of color quantization. This method introduces a probabilistic element to the quantization process, potentially improving efficiency and adaptability to diverse image characteristics. Results. To demonstrate the efficiency of our approach, we present experimental results using images from the ImageNet dataset. These experiments illustrate the performance of our Stochastic Quantization method in terms of compression quality, computational efficiency, and scalability compared to traditional color quantization techniques. Conclusions. This study introduces a scalable algorithm for solving the color quantization problem without memory constraints, demonstrating its efficiency on a subset of images from the ImageNet dataset. The convergence speed of the algorithm can be further enhanced by modifying the update rule with alternative methods to Stochastic Gradient Descent (SGD) that incorporate adaptive learning rates. Moreover, the stochastic nature of the proposed solution enables the utilization of parallelization techniques to simultaneously update the positions of multiple quants, potentially leading to significant performance improvements. This aspect of parallelization and its impact on algorithm efficiency presents a topic for future research. The proposed method not only addresses the limitations of existing color quantization techniques but also opens up new possibilities for optimizing image compression algorithms in resource-constrained environments. Keywords: non-convex optimization, stochastic optimization, stochastic quantization, color quantization, lossy compression.

Convolutional neural networks optimization using multi-objective particle swarm optimization algorithm

This paper presents a heuristic-based convolutional neural network architecture optimization scheme (HeuCNN). First, CNN architectures are encoded as particles in optimization space with two objectives test error and number of parameters. Then, the effectiveness of CNN particles is represented by a heuristic in optimization space. It is mathematically proven and experimentally shown that the proposed heuristic assigns the best values for CNNs with higher performances. Uniform and nonuniform distribution of CNN particles in optimization space are analyzed and behavior of CNNs moving towards higher test error and higher number of parameters are also mathematically proven. CNN architecture search is proposed based on particle swarm optimization (PSO) by new algorithms for particle difference and velocity computation as well as particle update rules. Limitations of HeuCNN are explained and primary insights are presented to address its challenges as future works. Evaluation of HeuCNN on 9 publicly available datasets and comparing with 38 constant and optimized CNN peer competitive models reveal that HeuCNN achieves the best test error and the lowest number of parameters simultaneously. Source codes, data and software associated with this research will be available online after the publication of the paper.

Weight-adaptive augmented Lagrange multiplier sequential convex programming for nonlinear trajectory optimization

Sequential convex programming has become a prominent research focus in aerospace trajectory optimization due to its high computational efficiency. One critical challenge in sequential convex programming is the infeasibility of subproblems, which significantly impacts its stability. To address this issue, certain constraints must be relaxed using relaxation variables, which are then incorporated into the objective function as a penalty term. However, the choice of penalty weight greatly influences the optimality and convergence of sequential convex programming. Therefore, this paper proposes a weight-adaptive augmented Lagrange multiplier sequential convex programming algorithm to tackle the challenge. First, this paper analyzes the difficulty of determining an appropriate penalty weight from the perspective of its penalty effect. Based on this analysis, an augmented penalty term is introduced, and the update rule for the augmented Lagrange multiplier is formulated using the Karush-Kuhn-Tucker conditions. Subsequently, a method with adaptive weight adjustment is designed based on the change in the penalty term to further improve the algorithm's solving speed. Finally, the proposed method is validated through numerical simulations in a Mars-powered landing problem and a small glide vehicle reentry problem. The results show that the proposed method is 20.59 % and 25.16 % faster compared to the general sequential convex programming algorithm, respectively. Furthermore, the proposed method is demonstrated to be robust to the initial choice of penalty weight.

Актуальные тенденции в законодательстве Китая о международном арбитраже

China did not adopt the UNCITRAL Model law in either version and its arbitration legislation often does not reflect the world-recognised practices of international arbitration. However, during the last few years, China has been softening its approach towards both international commercial and investment arbitration. The recent court decisions, adoption of the new Civil Procedure Law and proposal of the Draft Amended Arbitration Law signify gradual recognition of the concept of ‘seat of arbitration’ instead of relying on the domicile of the arbitration institution. Furthermore, Chinese courts have acknowledged the capacity of foreign arbitration institutions to administer disputes within mainland China — a significant shift from past practices where only Chinese arbitration commissions could administer arbitration proceedings in China under the Arbitration Law. This change is also reflected in the Draft Amended Arbitration Law. Another crucial legislative change is adoption of the Foreign State Immunity Law, which came into effect in January 2024. By enacting this law, China has shifted from absolute to restrictive immunity in disputes between a person or entity and a state. Thereby China allowed claims against foreign states arising out of commercial activities or international investment treaties to be considered in Chinese courts. Nevertheless, challenges remain in regards of recognition and enforcement of investment arbitral awards or conducting of investment arbitration in China. First, China’s reservation under the New York Convention 1958 limits recognition and enforcement of foreign awards to ‘commercial’ disputes, excluding investor-state disputes. Besides, it is discussed that China has not taken proper steps to recognise the ICSID awards as national ones for purposes of enforcement, even though it ratified the ICISD Convention. Furthermore, formally under the current Arbitration Law the investment arbitration is not possible. Finally, considering the latest legislative amendments, recent court practice and updated arbitration rules of Chinese arbitration institutions, it can be concluded that China has taken the pro-arbitration approach.

Pink-noise dynamics in an evolutionary game on a regular graph.

We consider a multiplayer prisoner's dilemma game on a square lattice and regular graphs based on the pairwise-Fermi update rule, and we obtain heatmaps of the fraction of cooperators and the correlation of neighboring pairs. In the heatmap, we find a mixed region where cooperators and defectors coexist, and the correlation between neighbors is significantly enhanced. Moreover, we observe pink-noise behavior in the mixed region, where the power spectrum can be fitted by a power-law function of frequency. We also find that the pink-noise behavior can be reproduced in a simple random-walk model. In particular, we propose a modified random-walk model which can reproduce not only the pink-noise behavior but also the deviation from it observed in a low-frequency region.

Multiagent Trust Region Policy Optimization.

We extend trust region policy optimization (TRPO) to cooperative multiagent reinforcement learning (MARL) for partially observable Markov games (POMGs). We show that the policy update rule in TRPO can be equivalently transformed into a distributed consensus optimization for networked agents when the agents' observation is sufficient. By using a local convexification and trust-region method, we propose a fully decentralized MARL algorithm based on a distributed alternating direction method of multipliers (ADMM). During training, agents only share local policy ratios with neighbors via a peer-to-peer communication network. Compared with traditional centralized training methods in MARL, the proposed algorithm does not need a control center to collect global information, such as global state, collective reward, or shared policy and value network parameters. Experiments on two cooperative environments demonstrate the effectiveness of the proposed method.