“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Acknowledgements: We would like to thank the following for contributions and suggestions: Bill Baxter, Brian Templeton, Christian Rishoj, Christian Schroppel Douglas L. Theobald, Esben Hoegh-Rasmussen, Glynne Casteel, Jan Larsen, Jun Bin Gao, Jurgen Struckmeier, Kamil Dedecius, Korbinian Strimmer, Lars Christiansen, Lars Kai Hansen, Leland Wilkinson, Liguo He, Loic Thibaut, Miguel Barao, Ole Winther, Pavel Sakov, Stephan Hattinger, Vasile Sima, Vincent Rabaud, Zhaoshui He. We would also like thank The Oticon Foundation for funding our PhD studies.

The Matrix Cookbook

The vision of next generation 5G wireless communications lies in providing very high data rates (typically of Gbps order), extremely low latency, manifold increase in base station capacity, and significant improvement in users’ perceived quality of service (QoS), compared to current 4G LTE networks. Ever increasing proliferation of smart devices, introduction of new emerging multimedia applications, together with an exponential rise in wireless data (multimedia) demand and usage is already creating a significant burden on existing cellular networks. 5G wireless systems, with improved data rates, capacity, latency, and QoS are expected to be the panacea of most of the current cellular networks’ problems. In this survey, we make an exhaustive review of wireless evolution toward 5G networks. We first discuss the new architectural changes associated with the radio access network (RAN) design, including air interfaces, smart antennas, cloud and heterogeneous RAN. Subsequently, we make an in-depth survey of underlying novel mm-wave physical layer technologies, encompassing new channel model estimation, directional antenna design, beamforming algorithms, and massive MIMO technologies. Next, the details of MAC layer protocols and multiplexing schemes needed to efficiently support this new physical layer are discussed. We also look into the killer applications, considered as the major driving force behind 5G. In order to understand the improved user experience, we provide highlights of new QoS, QoE, and SON features associated with the 5G evolution. For alleviating the increased network energy consumption and operating expenditure, we make a detail review on energy awareness and cost efficiency. As understanding the current status of 5G implementation is important for its eventual commercialization, we also discuss relevant field trials, drive tests, and simulation experiments. Finally, we point out major existing research issues and identify possible future research directions.

Next Generation 5G Wireless Networks: A Comprehensive Survey

This paper presents control and coordination algorithms for groups of vehicles. The focus is on autonomous vehicle networks performing distributed sensing tasks where each vehicle plays the role of a mobile tunable sensor. The paper proposes gradient descent algorithms for a class of utility functions which encode optimal coverage and sensing policies. The resulting closed-loop behavior is adaptive, distributed, asynchronous, and verifiably correct.

Coverage control for mobile sensing networks

This book provides a broad survey of models and efficient algorithms for Nonnegative Matrix Factorization (NMF) This includes NMFs various extensions and modifications, especially Nonnegative Tensor Factorizations (NTF) and Nonnegative Tucker Decompositions (NTD) NMF/NTF and their extensions are increasingly used as tools in signal and image processing, and data analysis, having garnered interest due to their capability to provide new insights and relevant information about the complex latent relationships in experimental data sets It is suggested that NMF can provide meaningful components with physical interpretations; for example, in bioinformatics, NMF and its extensions have been successfully applied to gene expression, sequence analysis, the functional characterization of genes, clustering and text mining As such, the authors focus on the algorithms that are most useful in practice, looking at the fastest, most robust, and suitable for large-scale models Key features: Acts as a single source reference guide to NMF, collating information that is widely dispersed in current literature, including the authors own recently developed techniques in the subject area Uses generalized cost functions such as Bregman, Alpha and Beta divergences, to present practical implementations of several types of robust algorithms, in particular Multiplicative, Alternating Least Squares, Projected Gradient and Quasi Newton algorithms Provides a comparative analysis of the different methods in order to identify approximation error and complexity Includes pseudo codes and optimized MATLAB source codes for almost all algorithms presented in the book The increasing interest in nonnegative matrix and tensor factorizations, as well as decompositions and sparse representation of data, will ensure that this book is essential reading for engineers, scientists, researchers, industry practitioners and graduate students across signal and image processing; neuroscience; data mining and data analysis; computer science; bioinformatics; speech processing; biomedical engineering; and multimedia

Nonnegative Matrix and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind Source Separation

This paper studies dynamic scheduling of a self-interested battery charging station that provides fully-charged batteries for electric vehicle (EV) battery swapping services. The charging station receives multi-type battery orders from the demand side, and it can refuse the orders or admit part of the orders according to current system states. If admitted, battery orders have to be served completely before predefined deadlines. Based on Lyapunov optimization framework, a dynamic scheduling approach is developed, which allows the charging station to observe real-time system states and make scheduling decisions in an online fashion. In theoretical analysis, the feasibility and suboptimality of the proposed approach are proven. Based on the analysis, the feasible ranges of algorithm parameters are derived, ensuring that battery orders can be completed before deadlines. In simulation, actual real-time electricity data is used. The results show that the proposed approach satisfies the deadline constraints and achieves higher profit than other benchmark approaches.

Dynamic Scheduling of Multi-Type Battery Charging Stations for EV Battery Swapping

Being important information of both network conditions and energy trading, distribution locational marginal prices (DLMPs) are greatly useful for distribution network (DN) management. The goal of this paper is to develop a market mechanism that motivates price-making agents to trade active and reactive power at DLMPs. According to the optimal power flow problem based on second-order cone (SOC) relaxation, we first design the market rules that allow agents to strategically submit proposals of price and power generation/consumption. The market rules induce a noncooperative game between the DN operator and distributed generators. We prove that DLMP pricing can be implemented at any generalized Nash equilibrium (GNE) of the game. We also show that economic properties of strong budget balance, individual rationality, and system cost minimization can be achieved at any GNE. A decentralized algorithm is developed to reach a GNE of the game. Further, based on the convex-concave procedure, an extension of the market mechanism is proposed to cope with the situations that SOC relaxation is inexact. In simulation, the proposed market mechanism is compared with other game-theoretic models, and the economic properties are verified.

Nash Mechanisms for Market Design Based on Distribution Locational Marginal Prices

This paper presents a novel iterative learning feedback control method for linear parabolic distributed parameter systems with multiple collocated piecewise observation. Multiple actuators and sensors distributed at the same position of the spatial domain are utilized to perform collocated piecewise control and measurement operations. The advantage of the proposed method is that it combines the iterative learning algorithm and feedback technique. Not only can it use the iterative learning algorithm to track the desired output trajectory, but also the feedback control approach can be utilized to achieve real-time online update. By utilizing integration by parts, triangle inequality, mean value theorem for integrals and Gronwall lemma, two sufficient conditions based on the inequality constraints for the convergence analysis of the tracking error system are presented. Some simulation experiments are provided to prove the effectiveness of the proposed method. 

Iterative learning feedback control for linear parabolic distributed parameter systems with multiple collocated piecewise observation

Distinguishing the importance of views plays a key role in multi-view learning as each view often contributes differently to a specific task. However, existing strategies generally attach importance to each view blindly from a data representation perspective, which often fails to identify low-quality views then leading to poor performance. In this paper, by establishing a link between labeled data and the importance of different views, we propose a semi-supervised label-driven auto-weighted strategy that evaluates the importance of views from a labeling perspective to avoid the negative impact of unimportant or low-quality views. Based on this strategy, we propose a transductive semi-supervised auto-weighted multi-view classification model that learns the labels of multi-view data under the more reasonable weights of the views. The model is decoupled into three small-scale sub-problems that can efficiently be optimized. The experimental results on classification tasks show that our proposed method achieves very promising classification accuracy at a lower computational cost compared to other related methods, and the weight change experiments show that our proposed strategy can distinguish view importance more accurately than other related strategies especially when low-quality views are involved in multi-view learning. 

A semi-supervised label-driven auto-weighted strategy for multi-view data classification

In this paper, we investigate the optimal output tracking problem for linear discrete-time systems with unknown dynamics using reinforcement learning and robust output regulation theory. This output tracking problem only allows to utilize the outputs of the reference system and the controlled system, rather than their states, and differs from most existing tracking results that depend on the state of the system. The optimal tracking problem is formulated into a linear quadratic regulation problem by proposing a family of dynamic discrete-time controllers. Then, it is shown that solving the output tracking problem is equivalent to solving output regulation equations, whose solution, however, requires the knowledge of the complete and accurate system dynamics. To remove such a requirement, an off-policy reinforcement learning algorithm is proposed using only the measured output data along the trajectories of the system and the reference output. By introducing re-expression error and analyzing the rank condition of the parameterization matrix, we ensure the uniqueness of the proposed RL based optimal control via output feedback.

Shengli Xie

Papers

Dynamic Scheduling of Multi-Type Battery Charging Stations for EV Battery Swapping

Nash Mechanisms for Market Design Based on Distribution Locational Marginal Prices

Iterative learning feedback control for linear parabolic distributed parameter systems with multiple collocated piecewise observation

A semi-supervised label-driven auto-weighted strategy for multi-view data classification

Robust Output Regulation and Reinforcement Learning-based Output Tracking Design for Unknown Linear Discrete-Time Systems