检索结果-南通市图书馆

维普期刊数据库

Skill enhancement learning with knowledge distillation

在线全文

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第8期67卷 206-220页

作者： Naijun LIU Fuchun SUN Bin FANG Huaping LIU department of Computer science and technology Tsinghua University

Skill learning through reinforcement learning has significantly progressed in recent years. However, it often struggles to efficiently find optimal or near-optimal policies due to the inherent trial-and-error exploration in reinforcement learning. Although algorithms have been proposed to enhance skill learning efficacy, there is still much room for improvement in terms of skill learning performance and training stability. In this paper, we propose an algorithm called skill enhancement learning with knowledge distillation(SELKD), which integrates multiple actors and multiple critics for skill learning. SELKD employs knowledge distillation to establish a mutual learning mechanism among actors. To mitigate critic overestimation bias,we introduce a novel target value calculation method. We also perform theoretical analysis to ensure the convergence of SELKD. Finally, experiments are conducted on several continuous control tasks, illustrating the effectiveness of the proposed algorithm.

关键词： skill learning enhancement learning reinforcement learning knowledge distillation

同方期刊数据库

Topologically Correct Intersection Curves of Two Trimmed Quadrics with Tolerance Control

在线全文

学校读者我要写书评

暂无评论

Journal of Systems science & Complexity 2024年第5期37卷 2207-2239页

作者： SHAO Wenbing CHEN Falai department of Mathematics University of Science and Technology of China

Surface/surface intersection is a fundamental problem in Compute Aided Design and Geometric Modeling since it is essential to solid modeling, numerically controlled machining, feature recognition, computer animation, etc. In practical applications, quadric surfaces, which are the most basic type of surfaces, are typically bounded surfaces trimmed by a sequence of planes. In this paper,a robust algorithm is proposed for computing the intersection curve segments of two trimmed quadrics based on the parametric representation of the intersection curves of two quadrics. The proposed algorithm guarantees correct topology and ensures that the approximation errors of the end points of the intersection curve segments are less than a given tolerance. The error control is based on an effective solution to a set of polynomial inequality system using the root isolation technique. Some examples are presented to validate the robustness and effectiveness of the proposed algorithm.

关键词： Polynomial inequality system root isolation surface/surface intersection tolerance control trimmed quadrics

Significantly Improved High-Temperature Energy Storage Performance of BOPP Films by Coating Nanoscale Inorganic Layer

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Energy & Environmental Materials 2024年第2期7卷 34-42页

作者： Tiandong Zhang Hainan Yu Young Hoon Jung Changhai Zhang Yu Feng Qingguo Chen Keon Jae Lee Qingguo Chi School of Electrical and Electronic Engineering Harbin University of Science and Technology department of Materials science and Engineering Korea Advanced Institute of Science and Technology

Biaxially oriented polypropylene（BOPP） is one of the most commonly used commercial capacitor films, but its upper operating temperature is below105 °C due to the sharply increased electrical conduction loss at high temperature. In this study, growing an inorganic nanoscale coating layer onto the BOPP film's surface is proposed to suppress electrical conduction loss at high temperature, as well as increase its upper operating *** kinds of inorganic coating layers that have different energy band structure and dielectric property are grown onto the both surface of BOPP films, respectively. The effect of inorganic coating layer on the hightemperature energy storage performance has been systematically investigated. The favorable coating layer materials and appropriate thickness enable the BOPP films to have a significant improvement in hightemperature energy storage performance. Specifically, when the aluminum nitride（Al N） acts as a coating layer, the Al N-BOPP-Al N sandwich-structured films possess a discharged energy density of 1.5 J cm-3with an efficiency of90% at 125 °C, accompanying an outstandingly cyclic property. Both the discharged energy density and operation temperature are significantly enhanced, indicating that this efficient and facile method provides an important reference to improve the high-temperature energy storage performance of polymer-based dielectric films.

关键词： coating layer energy storage interfacial barrier polymer films

Physics-informed deep Koopman operator for Lagrangian dynamic systems

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第09期67卷 146-163页

作者： Xuefeng WANG Yang CAO Shaofeng CHEN Yu KANG department of Automation University of Science and Technology of China State Key Laboratory of Fire science University of Science and Technology of China Institute of Advanced technology University of Science and Technology of China

Accurate mechanical system models are crucial for safe and stable control. Unlike linear systems,Lagrangian systems are highly nonlinear and difficult to optimize because of their unknown system *** research thus used deep neural networks to generate linear models of original systems by mapping nonlinear dynamic systems into a linear space with a Koopman observable function encoder. The controller then relies on the Koopman linear model. However, without physical information constraints, ensuring control consistency between the original nonlinear system and the Koopman system is tough, as the learning process of the Koopman observation function is unsupervised. This paper thus proposes a two-stage learning algorithm that uses structural subnetworks to build a physics-informed network topology to simultaneously learn the Koopman observable functions and the system energy representation. In the Koopman matrix learning session, a quadratic-constrained optimization problem is solved to ensure that the Koopman representation satisfies the energy difference matching hard constraint. The proposed energy-preserving deep Lagrangian Koopman(EPDLK) framework effectively represents the dynamics of the Lagrangian system while ensuring control consistency. The effectiveness of EPDLK is compared with those of various Koopman observable function construction methods in multistep prediction and trajectory tracking tasks. EPDLK achieves better control consistency by guaranteeing energy difference matching, which facilitates the application of the control law generated on the Koopman system directly to the original nonlinear Lagrangian system.

关键词： Koopman operator Lagrangian physics-informed nonlinear model predictive control (MPC)

同方期刊数据库

Berry-Esseen bounds for self-normalized sums of locally dependent random variables

在线全文

学校读者我要写书评

暂无评论

science China Mathematics 2024年第11期67卷 2629-2652页

作者： Zhuo-Song Zhang department of Statistics and Data science Southern University of Science and Technology

The Berry-Esseen bound provides an upper bound on the Kolmogorov distance between a random variable and the normal *** this paper,we establish Berry-Esseen bounds with optimal rates for self-normalized sums of locally dependent random variables,assuming only a second-moment *** proof leverages Stein's method and introduces a novel randomized concentration inequality,which may also be of independent interest for other *** main results have applied to self-normalized sums of m-dependent random variables and graph dependency models.

关键词： Berry-Esseen bounds self-normalized sums localdependence $m$-dependence graph dependency

同方期刊数据库

Homovalent doping: An efficient strategy of the enhanced TiNb2O7 anode for lithium-ion batteries

在线全文

学校读者我要写书评

暂无评论

Green Energy & Environment 2024年第8期9卷 1257-1266页

作者： Xiaohe Jin Yirui Deng Han Tian Miaomiao Zhou Wenhao Tang Huiyou Dong Xinquan Zhang Ruiping Liu department of Materials science and Engineering China University of Mining & Technology (Beijing) department of Materials science and Engineering China University of Mining&Technology (Beijing)

The low energy density,unsatisfied cycling performance,potential safety issue and slow charging kinetics of the commercial lithium-ion batteries restrained their further application in the fields of fast charging and long-haul electric *** TiNb2O7（TNO） with the theoretical capacity of 387 mAh g-1has been proposed as a high-capacity anode materials to replace *** this work,homovalent doping strategy was used to enhance the electrochemical performance of Ti Nb2O7（TNO） by employing Zr to partial substitute Ti through solvothermal *** doping of Zr4+ions can enlarge the lattice structure without changing the chemical valence of the original elements,refine and homogenize the grains,improve the electrical conductivity,and accelerate the ion diffusion kinetics,and finally enhance the cycle and rate ***,Z0.05-TNO shows initial discharge capacity of as high as 312.2 mAh g-1at 1 C and 244.8 mAh g-1at 10 C,and still maintains a high specific capacity of 171.3 mAh g-1after 800 cycles at 10 *** study provides a new strategy for high-performance fastcharging energy storage electrodes.

关键词： Homovalent doping Zr 4+ TiNb 2 O 7 Microsphere Lithium-ion batteries

Rotational Doppler effect of composite vortex beams with tailored OAM spectra

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Photonics Research 2024年第8期12卷 1665-1672页

作者： YUTIAN LIANG RUIJIAN LI JIE ZHAO XINGYUAN LU TONG LIU ZHENGLIANG LIU YUAN REN CHENGLIANG ZHAO department of Aerospace science and technology Space Engineering University School of Physical science and technology Soochow University department of Basic Course Space Engineering University

There recently has been increasing interest in the research and application of the rotational Doppler effect(RDE),which paves a promising way to detect rotating objects remotely. In order to obtain more information about the rotating object from the rotational Doppler signal, composite vortex beams by coaxial superposition of orbital angular momentum(OAM) modes are often used as the probe beam. However, to the best of our knowledge, the RDE of composite vortex beams with arbitrary OAM spectra has not yet been comprehensively studied. In this paper, the correspondence between the OAM spectrum of a probe beam and the frequency spectrum of a rotational Doppler signal is theoretically analyzed. It is explicitly revealed that the RDE frequency spectrum of scattered light is related to the product of two autocorrelation functions: one from the OAM spectrum of probe beam and the other from the spiral spectrum of rotating object. On the basis of this relation, one can regulate the RDE frequency spectrum on demand via tailoring the OAM spectrum of the probe beam. As a proof of concept we design a special composite vortex beam to eliminate the broadening of the RDE spectrum induced by misalignment. These findings are of practical value in applications such as remote sensing and optical metrology.

关键词： Circular polarization Doppler effect Light beams Optical vortex beams Remote sensing Spatial light modulators

Federated Learning Security and Privacy-Preserving Algorithm and Experiments Research Under Internet of Things Critical Infrastructure

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Tsinghua science and technology 2024年第2期29卷 400-414页

作者： Nasir Ahmad Jalali Hongsong Chen department of Computer science University of Science and Technology Beijing(USTB)Beijing 100083China

The widespread use of the Internet of Things(IoTs)and the rapid development of artificial intelligence technologies have enabled applications to cross commercial and industrial band *** such systems,all participants related to commercial and industrial systems must communicate and generate ***,due to the small storage capacities of IoT devices,they are required to store and transfer the generated data to third-party entity called“cloud”,which creates one single point to store their ***,as the number of participants increases,the size of generated data also ***,such a centralized mechanism for data collection and exchange between participants is likely to face numerous challenges in terms of security,privacy,and *** address these challenges,Federated Learning(FL)has been proposed as a reasonable decentralizing approach,in which clients no longer need to transfer and store real data in the central ***,they only share updated training models that are trained over their private *** the same time,FL enables clients in distributed systems to share their machine learning models collaboratively without their training data,thus reducing data privacy and security ***,slow model training and the execution of additional unnecessary communication rounds may hinder FL applications from operating properly in a distributed ***,these unnecessary communication rounds make the system vulnerable to security and privacy issues,because irrelevant model updates are sent between clients and ***,in this work,we propose an algorithm for fully homomorphic encryption called Cheon-Kim-Kim-Song(CKKS)to encrypt model parameters for their local information privacy-preserving *** proposed solution uses the impetus term to speed up model convergence during the model training ***,it establishes a secure communication channel between IoT devices and the *** a

关键词： Federated Learning(FL) Internet of Things(IoTs) lightweight Transport Layer Security(iTLS) Cheon-Kim-Kim-Song(CKKS)

Memory-enhanced text style transfer with dynamic style learning and calibration

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第4期67卷 181-196页

作者： Fuqiang LIN Yiping SONG Zhiliang TIAN Wangqun CHEN Diwen DONG Bo LIU College of Computer science and technology National University of Defense Technology department of Computer science and Engineering The Hong Kong University of Science and Technology Strategic Assessments and Consultation Institute Academy of Military Sciences

Text style transfer aims to rephrase a sentence to match the desired style while retaining the original content. As a controllable text generation task, mainstream approaches use content-independent style embedding as control variables to guide stylistic generation. Nonetheless, stylistic properties are contextsensitive even under the same style. For example, “delicious” and “helpful” convey positive sentiments,although they are more likely to describe food and people, respectively. Therefore, desired style signals must vary with the content. To this end, we propose a memory-enhanced transfer method, which learns fine-grained style representation concerning content to assist transfer. Rather than employing static style embedding or latent variables, our method abstracts linguistic characteristics from training corpora and memorizes subdivided content with the corresponding style representations. The style signal is dynamically retrieved from memory using the content as a query, providing a more expressive and flexible latent style space. To address the imbalance between quantity and quality in different content, we further introduce a calibration method to augment memory construction by modeling the relationship between candidate *** results obtained using three benchmark datasets confirm the superior performance of our model compared to competitive approaches. The evaluation metrics and case study also indicate that our model can generate diverse stylistic phrases matching context.

关键词： style transfer memory-enhanced method text generation deep learning text representation