2021 55th Annual Conference on Information Sciences and Systems (CISS)最新文献

{"title":"Competitive Control via Online Optimization with Memory, Delayed Feedback, and Inexact Predictions","authors":"Guanya Shi","doi":"10.1109/CISS50987.2021.9400281","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400281","url":null,"abstract":"Recently a line of work has shown the applicability of tools from online optimization for control, leading to online control algorithms with learning-theoretic guarantees, such as sublinear regret. However, the predominant benchmark, static regret, only compares to the best static linear controller in hindsight, which could be arbitrarily sub-optimal compared to the true offline optimal policy in non-stationary environments. Moreover, the common robustness considerations in control theory literature, such as feedback delays and inexact predictions, only have little progress in the context of online learning/optimization guarantees. In this talk, based on our three recent papers, I will present key principles and practical algorithms towards online control with competitive ratio guarantees, which directly bound the suboptimality compared to the true offline optimal policy. First, I will show the deep connections between a novel class of online optimization with memory and online control, which directly translates online optimization guarantees to online control guarantees and gives the first constant-competitive policy with adversarial disturbances [1]. Second, I will analyze the performance of the most popular online policy in the control community, Model Predictive Control (MPC), from the online learning's perspective, and show a few important fundamental limits. Our results give the first finite-time performance guarantees for MPC [3]. Finally, I will discuss the influence of delayed feedback and inexact predictions on competitive ratio analysis [2].","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116374456","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}

{"title":"A PAC-bound on the Channel Capacity of an Observed Discrete Memoryless Channel","authors":"M. A. Tope, Joel M. Morris","doi":"10.1109/CISS50987.2021.9400323","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400323","url":null,"abstract":"This paper presents a method to compute the channel capacity of an observed (partially known) discrete memoryless channel (DMC) using a probably approximately correct (PAC) bound. Given $N$ independently and identically distributed (i.i.d.) input-output sample pairs, we define a compound DMC with convex sublevel-sets to constrain the channel output uncertainty with high probability. Then we numerically solve an ‘K-way’ convex optimization to determine an achievable information rate $R_{L}(N)$ across the channel that holds with a specified high probability. Our approach provides the non-asymptotic ‘worst-case’ convergence $R_{L}(N)$ to channel capacity $C$ at the rate of $O(sqrt{log (log (N)) / N})$.","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"303 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123240812","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}

{"title":"Optimal VWAP Strategies under Regime Switching","authors":"M. Pemy","doi":"10.1109/CISS50987.2021.9400284","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400284","url":null,"abstract":"In this paper, we analyze the problem of trading a large position in the market place when the stock price dynamic follows a regime switching process. In this work, we are particularly interested in trading algorithms that track the market benchmark known as the volume-weighted average price (VWAP). We propose a trading algorithm that breaks the execution order into small pieces and executes them over a predetermined period of time so as to maximize its VWAP or possibly exceed the overall market VWAP. The underlying problem is formulated as a discrete-time stochastic optimal control problem with resource constraints. The value function and optimal trading strategies are derived in closed-form. Numerical simulations with market data are reported to illustrate the pertinence of these results.","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127481166","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}

{"title":"Energy Efficiency Analysis of RLS-MUSIC based Smart Antenna System for 5G Network","authors":"H. B. Kassa, K. Kornegay, Ebrima N. Ceesay","doi":"10.1109/CISS50987.2021.9400325","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400325","url":null,"abstract":"This paper proposes a novel energy efficient beamforming and DOA estimation scheme called RLS-MUSIC algorithm. First, we study different adaptive beamforming algorithms and receive signal Degree-of-Arrival (DOA) estimation techniques for wireless communication network application in reference to their effect on energy efficiency. Taking linear array smart antenna elements geometry, which is mostly used antenna array geometry, Recursive Least Square (RLS) adaptive beam-forming algorithm and Multiple Signal Classification (MUSIC) DOA estimation technique are identified as the reference input methods to model the new energy efficient scheme. Using an energy model, which is as a function of beam-width, signal transmission strength (range), and signal-to-noise-interference ratio (SNIR), we evaluate the total amount of energy spent during beam-formation. Finally, the energy gain due to RLS-MUSIC scheme is compared with that of RLS alone. The results show that the RLS-MUSIC scheme based access-points/base-station antenna systems consume less energy during beam formation and they generate beams with narrower beam-width, lower side lobe level and better deep null.","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128554132","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}

{"title":"Machine Learning Models for Human Fall Detection using Millimeter Wave Sensor","authors":"Mubarak A. Alanazi, Abdullah K. Alhazmi, C. Yakopcic, V. Chodavarapu","doi":"10.1109/CISS50987.2021.9400259","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400259","url":null,"abstract":"Accidental falls are a common threat to the health of older adults, which can reduce their ability to remain independent. Fall detection sensors have become essential lifesaving health monitoring systems for the elderly. We describe a privacy protecting system for stance monitoring of occupants within a room using a millimeter wave (mmWave) sensor. We studied various machine learning models that are best suited to analyze the response from a mmWave system output. After comparing several machine learning algorithms, we found that feedforward neural networks provide the highest test accuracy.","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"72 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130822836","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}

{"title":"Human Activity Recognition Through Ensemble Learning of Multiple Convolutional Neural Networks","authors":"Narjis Zehra, Syed Hamza Azeem, M. Farhan","doi":"10.1109/CISS50987.2021.9400290","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400290","url":null,"abstract":"Human Activity Recognition is a field concerned with the recognition of physical human activities based on the interpretation of sensor data, including one-dimensional time series data. Traditionally, hand-crafted features are relied upon to develop the machine learning models for activity recognition. However, that is a challenging task and requires a high degree of domain expertise and feature engineering. With the development in deep neural networks, it is much easier as models can automatically learn features from raw sensor data, yielding improved classification results. In this paper, we present a novel approach for human activity recognition using ensemble learning of multiple convolutional neural network (CNN) models. Three different CNN models are trained on the publicly available dataset and multiple ensembles of the models are created. The ensemble of the first two models gives an accuracy of 94% which is better than the methods available in the literature.","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125455857","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}

{"title":"Controlling Collective Motions of Self-Propelled Particles by Mean Field Couplings Defined by Topology","authors":"T. Woolf, Justin Tervala, I. Carter","doi":"10.1109/CISS50987.2021.9400320","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400320","url":null,"abstract":"Weakly interacting particle systems have been important model systems for exploring the behavior of flocks of birds, of swarms of fish, and of human crowds and traffic patterns. Models from the microscopic level of the individual have focused on finding parameters that when time-evolved show stable states. Recent work on the D'Orsogna model showed that topological analysis can be used more effectively than order parameters to define the stable states that are determined by collective motions. Using ideas from Mean Field Games and Mean Field Controls we show that a coupling to the individual behaviors can be added to control collective states defined previously by analysis of trajectory data. This leads us to define controls for interacting systems where the macroscopic couplings are defined by topological descriptors. Our approach extends work on traffic flows that used one-dimensional models and density estimates to couple microscopic and macroscopic behavior. This coupling between microscopic and macroscopic behavior defined by topological analysis may be fruitful for control of continuous dynamical systems, often described by non-linear coupled differential equations, where predefined sets of finite states are often difficult to describe in advance. We thus suggest that controls defined in this manner can be a form of dual control, where collective motions are defined and controlled without a need to understand the entire event space from microscopic details.","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123340917","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}

{"title":"Cache Freshness in Information Updating Systems","authors":"Melih Bastopcu, S. Ulukus","doi":"10.1109/CISS50987.2021.9400310","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400310","url":null,"abstract":"We consider a cache updating system with a source, <tex>$m$</tex> caches and a user. There are <tex>$n$</tex> files. The source keeps the freshest version of the files which are updated with known rates <tex>$lambda_{i}$</tex>. The first cache downloads and keeps the freshest version of the files from the source with rates <tex>$c_{1i}$</tex> and cache <tex>$r$</tex> downloads and keeps the files from cache <tex>$r-1$</tex> with rates <tex>$c_{ri}$</tex> for <tex>$r > 1$</tex>. The user gets updates from cache <tex>$m$</tex> with rates <tex>$u_{i}$</tex>. When the user gets an update, it either gets a fresh update from cache <tex>$m$</tex> or the file at cache <tex>$m$</tex> becomes outdated by a file update at the source. We find an analytical expression for the average freshness of the files at the user. We provide an alternating maximization based method to find the update rates for the caches, <tex>$c_{ri}$</tex>, and for the user, <tex>$u_{i}$</tex>, to maximize the freshness of the files at the user. We note that for a given set of update rates for the user (resp. for the caches), the optimal rate allocation policy for the caches (resp. for the user) is a threshold policy, where the optimal update rates for rapidly changing files at the source may be equal to zero.","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116955198","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}

{"title":"Approaching Prosumer Social Optimum via Energy Sharing with Proof of Convergence","authors":"Yue-Chun Chen, Changhong Zhao, S. Low, S. Mei","doi":"10.1109/CISS50987.2021.9400276","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400276","url":null,"abstract":"The conventional power system operation approach has been proven to be effective and reliable for decades. Specially, at the demand side, customers are managed centrally by aggregators and usually not price-responsive. With the prevalence of distributed energy resources (DERs), traditional consumers are now endowed with the ability to produce energy, turning into so-called prosumers. Prosumers can tradeoff between supply and demand and participate in energy management proactively. At the same time, the intermittent and uncertain nature of DERs call for a stronger capability of dealing with real-time energy fluctuation. In this context, exploiting demand-side flexibility to support real-time energy balancing, which can reduce required generation reserves and save costs, is a promising direction for energy system modernization. However, the traditional centralized scheme fails to allow a prosumer to act upon its profit-maximizing philosophy, which reduces prosumer incentives and restricts demand-side flexibility. Therefore, a new prosumer-oriented approach is desired. The emerging sharing economy in recent years, such as Airbnb and Uber, is a successful business model to enhance resource utilization efficiency by encouraging exchanges of goods among customers. Inspired by this, it is natural to ask: can energy be shared similarly? That is, is it possible to motivate idle generation resources to help support real-time energy balancing instead of relying on additional reserves? In this paper, an energy sharing mechanism is presented to achieve this goal. The main contributions are three-fold: 1) Mechanism Design. An energy sharing mechanism is proposed to accommodate prosumers' strategic decision-making on their self-production and demand in the presence of capacity constraints. The relationship between sharing quantity and sharing price is described by a generic supply-demand function so that prosumers' market roles are endogenously determined. Under this setting, all prosumers constitute a generalized Nash game whose equilibrium is difficult to characterize. Moreover, the inequality capacity constraints will further complicate equilibrium analysis. 2) Equilibrium. Main properties of the proposed energy sharing game are proved. The generalized Nash equilibrium of the game always exists and is partially unique. No prosumer is worse off so that it always has the incentive to participate in energy sharing. The energy sharing game achieves a $1-O(1/I)$ price-of-anarchy (PoA, which is less than 1 because the net cost is negative in our setting). Besides, as the number of prosumers $I$ increases, both the total net cost and individual prosumer strategies are proved to converge to the social optimum under the centralized operation. 3) Algorithm. A bidding process is developed to achieve the desired equilibrium in a distributed manner. This paper provides guidance for selecting the market sensitivity parameter so that the bidding process is guaranteed to co","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125723604","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}

{"title":"SAFE LINEAR BANDITS","authors":"Ahmadreza Moradipari, Sanae Amani, M. Alizadeh, Christos Thrampoulidis","doi":"10.1109/CISS50987.2021.9400288","DOIUrl":"https://doi.org/10.1109/CISS50987.2021.9400288","url":null,"abstract":"Bandit algorithms have various applications in safety-critical systems, where it is important to respect the system's underlying constraints. The challenge is that such constraints are often unknown as they depend on the bandit's unknown parameters. In this talk, we formulate a linear stochastic multi-armed bandit problem with safety constraints that depend linearly on an unknown parameter vector. As such, the learner is unable to identify all safe actions and must act conservatively in ensuring that their actions satisfy the safety constraint at all rounds (at least with high probability). For these bandits, we propose new upper-confidence bound (UCB) and Thompson-sampling algorithms, which include necessary modifications to respect the safety constraints. For two settings -with and without bandit feedback information on the constraint- we prove regret bounds and discuss their optimality in relation to corresponding bounds in the absence of safety restrictions. For example, for a setting with bandit-feedback information on the constraint, we present a frequentist regret of order $mathcal{O}left(d^{3/2}log^{1/2}dsqrt{T}log^{2/3}Tright)$, which remarkably matches the results provided by [1] for the standard linear Thompson-sampling algorithm. We highlight how the inherently randomized nature of Thompson-sampling helps expand the set of safe actions the algorithm has access to at each round. Finally, we discuss related problem variations with stage-wise baseline constraints, in which the learner must choose actions that not only maximize cumulative reward across the entire time horizon, but they further satisfy a linear baseline constraint taking the form of a lower bound on the instantaneous reward. The content of this talk is based on [2]–[4].","PeriodicalId":228112,"journal":{"name":"2021 55th Annual Conference on Information Sciences and Systems (CISS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129231500","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}