IEEE Control Systems Letters最新文献_第10页

We Are Legion: High Probability Regret Bound in Adversarial Multiagent Online Learning 我们是军团：对抗性多智能体在线学习中的高概率后悔边界

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-18 DOI: 10.1109/LCSYS.2024.3519637

Sri Jaladi;Ilai Bistritz

{"title":"We Are Legion: High Probability Regret Bound in Adversarial Multiagent Online Learning","authors":"Sri Jaladi;Ilai Bistritz","doi":"10.1109/LCSYS.2024.3519637","DOIUrl":"https://doi.org/10.1109/LCSYS.2024.3519637","url":null,"abstract":"We study a large-scale multiagent online learning problem where the number of agents N is significantly larger than the number of arms K. The agents face the same adversarial online learning problem with K arms over T rounds, where the adversary chooses the cost vectors <inline-formula> <tex-math>$boldsymbol {l}(1), ldots ,boldsymbol {l}(T)$ </tex-math></inline-formula> before the game begins. Each round t, each agent n picks an arm <inline-formula> <tex-math>$a_{n}$ </tex-math></inline-formula>(t) and incurs a cost of <inline-formula> <tex-math>$l_{a_{n}(t)}$ </tex-math></inline-formula> (t). Then, at the end of the round, all agents observe the costs of all arms <inline-formula> <tex-math>$l_{1}(t), ldots ,l_{K}(t)$ </tex-math></inline-formula>. The exponential weights algorithm achieves an order-wise optimal expected regret of <inline-formula> <tex-math>$O(sqrt {T})$ </tex-math></inline-formula> for each agent. However, the variance of the sum of regrets scales linearly with the number of agents, which is unacceptable for a large-scale multi-agent system. To mitigate this, we propose a simple fully distributed algorithm that achieves the same optimal expected sum of regrets but reduces the variance of the sum of regrets from O(N) to <inline-formula> <tex-math>$O(min (N,K))$ </tex-math></inline-formula> with no communication required between the agents.","PeriodicalId":37235,"journal":{"name":"IEEE Control Systems Letters","volume":"8 ","pages":"2985-2990"},"PeriodicalIF":2.4,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142975927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fast-and-Flexible Decision-Making With Modulatory Interactions

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-17 DOI: 10.1109/LCSYS.2024.3519445

Rodrigo Moreno-Morton;Anastasia Bizyaeva;Naomi Ehrich Leonard;Alessio Franci

引用次数: 0

Data-Driven Predictive Control of Bilinear HVAC Dynamics—An Experimental Case Study 数据驱动的双线性HVAC动态预测控制——一个实验案例研究

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-17 DOI: 10.1109/LCSYS.2024.3519224

Deborah Bilgic;Alexander Harding;Timm Faulwasser

引用次数: 0

Adversarial Sensor Attacks Against Uncertain Cyber-Physical Systems: A Dynamic Output Feedback Approach 针对不确定网络物理系统的对抗性传感器攻击：一种动态输出反馈方法

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-17 DOI: 10.1109/LCSYS.2024.3519435

Kartik A. Pant;Shiraz Khan;Inseok Hwang

引用次数: 0

On Generating Explanations for Reinforcement Learning Policies: An Empirical Study 关于生成强化学习策略解释的实证研究

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-17 DOI: 10.1109/LCSYS.2024.3519301

Mikihisa Yuasa;Huy T. Tran;Ramavarapu S. Sreenivas

引用次数: 0

A New Semi-Discretization of the Fully Clamped Euler-Bernoulli Beam Preserving Boundary Observability Uniformly 全钳位欧拉-伯努利梁的新半离散化，均匀保留边界可观测性

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-17 DOI: 10.1109/LCSYS.2024.3519379

Ahmet Kaan Aydin;Md Zulfiqur Haider;Ahmet Özkan Özer

{"title":"A New Semi-Discretization of the Fully Clamped Euler-Bernoulli Beam Preserving Boundary Observability Uniformly","authors":"Ahmet Kaan Aydin;Md Zulfiqur Haider;Ahmet Özkan Özer","doi":"10.1109/LCSYS.2024.3519379","DOIUrl":"https://doi.org/10.1109/LCSYS.2024.3519379","url":null,"abstract":"This letter extends a Finite Difference model reduction method to the Euler-Bernoulli beam equation with fully clamped boundary conditions. The corresponding partial differential equation (PDE) is exactly observable in the energy space with a single boundary observer in arbitrarily short observation times. However, standard Finite Difference spatial discretization fails to achieve uniform exact observability as the mesh parameter approaches zero, with minimal observation time potentially depending on the filtering parameter. To address this, we propose a Finite Difference algorithm incorporating an averaging operator and discrete multipliers, leveraging Haraux’s theorem on the spectral gap to ensure uniform observability. This approach eliminates the need for artificial viscosity or Fourier filtering. Our method achieves uniform observability for arbitrarily small times with dual observers-the tip moment and average tip velocity-mirroring results from mixed Finite Elements applied to the wave equation with homogeneous Dirichlet boundary conditions, where dual controllers converge to the single controller of the PDE model [Castro, Micu-Numerische Mathematik’06]. Our reduced model is applicable to more complex systems involving Euler-Bernoulli beam equations.","PeriodicalId":37235,"journal":{"name":"IEEE Control Systems Letters","volume":"8 ","pages":"2955-2960"},"PeriodicalIF":2.4,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142975846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

On the O(1/k) Convergence of Distributed Gradient Methods Under Random Quantization 随机量化下分布梯度方法的O（1/k）收敛性

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-16 DOI: 10.1109/LCSYS.2024.3519013

Amit Dutta;Thinh T. Doan

{"title":"On the O(1/k) Convergence of Distributed Gradient Methods Under Random Quantization","authors":"Amit Dutta;Thinh T. Doan","doi":"10.1109/LCSYS.2024.3519013","DOIUrl":"https://doi.org/10.1109/LCSYS.2024.3519013","url":null,"abstract":"We revisit the so-called distributed two-time-scale stochastic gradient method for solving a strongly convex optimization problem over a network of agents in a bandwidth-limited regime. In this setting, the agents can only exchange the quantized values of their local variables using a limited number of communication bits. Due to quantization errors, the existing best-known convergence results of this method can only achieve a suboptimal rate <inline-formula> <tex-math>$mathcal {O}$ </tex-math></inline-formula>(<inline-formula> <tex-math>$1/sqrt {k}$ </tex-math></inline-formula>), while the optimal rate is <inline-formula> <tex-math>$mathcal {O}$ </tex-math></inline-formula>(<inline-formula> <tex-math>$1/k$ </tex-math></inline-formula>) under no quantization, where k is the time iteration. The main contribution of this letter is to address this theoretical gap, where we study a sufficient condition and develop an innovative analysis and step-size selection to achieve the optimal convergence rate <inline-formula> <tex-math>$mathcal {O}$ </tex-math></inline-formula>(<inline-formula> <tex-math>$1/k$ </tex-math></inline-formula>) for the distributed gradient methods given any number of quantization bits. We provide numerical simulations to illustrate the effectiveness of our theoretical results.","PeriodicalId":37235,"journal":{"name":"IEEE Control Systems Letters","volume":"8 ","pages":"2967-2972"},"PeriodicalIF":2.4,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142975932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

From a Single Trajectory to Safety Controller Synthesis of Discrete-Time Nonlinear Polynomial Systems 从单轨迹到安全控制器的离散时间非线性多项式系统综合

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-16 DOI: 10.1109/LCSYS.2024.3519017

Behrad Samari;Omid Akbarzadeh;Mahdieh Zaker;Abolfazl Lavaei

引用次数: 0

Finite-Dimensional Observer-Based Boundary Control of 1-D Linear Parabolic-Elliptic Systems 一维线性抛物-椭圆系统的有限维观测器边界控制

IF 2.4

IEEE Control Systems Letters Pub Date : 2024-12-16 DOI: 10.1109/LCSYS.2024.3518396

Pengfei Wang;Emilia Fridman

引用次数: 0

Investigating Bistable Dynamics of Coupled Oscillators With Similarities to Neural Activity in Epilepsy 研究与癫痫神经活动相似的耦合振荡器的双稳态动力学