检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Borsdorf, Marvin Pan, Zexu Himmelmann, Pascal Li, Haizhou Schultz, Tanja Machine Listening Lab University of Bremen Bremen Germany Alibaba Group Singapore Department of Mathematics and Computer Science FernUniversität in Hagen Hagen Germany School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China Department of Electrical and Computer Engineering National University of Singapore Singapore Cognitive Systems Lab University of Bremen Bremen Germany

ISBN: (纸本)9798350368741

Speech separation aims to equip machines with the human ability of selective listening, i.e. to focus attention on specific information in spoken communication. Studies have shown that the language spoken in a cocktail party scenario matters. While the development of speech separation models can leverage extensive databases, for the majority of languages only very limited data is available. This work presents the very first study on speech separation for low-resource languages. We choose blind source separation as the task to be studied and analyze three strategies to overcome the data scarcity of two low-resource languages from the GlobalPhoneMS2 database. We show that data from other languages can be used to develop models that work for low-resource languages. Finetuning additionally boosts the performance, and training on multiple languages increases both performance and robustness. We show that dynamic mixing in the development helps to find a trade-off between performance and development time. © 2025 IEEE.

关键词： cocktail party problem GlobalPhone low-resource language multilingual speech separation

来源：评论

学校读者我要写书评

暂无评论

Convergence of energy-based learning in linear resistive networks

arXiv

引用

arXiv 2025年

作者： Huijzer, Anne-Men Chaffey, Thomas Besselink, Bart van Waarde, Henk J. Bernoulli Institute for Mathematics Computer Science and Artificial Intelligence University of Groningen Groningen Netherlands CogniGron - Groningen Cognitive Systems and Materials Center Netherlands Control Group Department of Engineering University of Cambridge United Kingdom School of Electrical and Computer Engineering University of Sydney Australia

Energy-based learning algorithms are alternatives to backpropagation and are well-suited to distributed implementations in analog electronic devices. However, a rigorous theory of convergence is lacking. We make a first step in this direction by analysing a particular energy-based learning algorithm, Contrastive Learning, applied to a network of linear adjustable resistors. It is shown that, in this setup, Contrastive Learning is equivalent to projected gradient descent on a convex function, for any step size, giving a guarantee of convergence for the *** Codes 65K10, 68T05, 93B30, 93D99 © 2025, CC BY-SA.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Speech Separation for Low-Resource Languages

Speech Separation for Low-Resource Languages

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Marvin Borsdorf Zexu Pan Pascal Himmelmann Haizhou Li Tanja Schultz Machine Listening Lab University of Bremen Bremen Germany Alibaba Group Singapore Department of Mathematics and Computer Science FernUniversität in Hagen Hagen Germany School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China Department of Electrical and Computer Engineering National University of Singapore Singapore Cognitive Systems Lab University of Bremen Bremen Germany

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

关键词： Training Measurement Databases Signal processing Data models Blind source separation Robustness Acoustics Speech processing

来源：评论

学校读者我要写书评

暂无评论

Pricing and Inventory Control Strategy for a Periodic-Review Energy Buy-Back System

引用

Journal of systems science & Complexity 2016年第4期29卷 1018-1033页

作者： ZHANG Jihong CHEN Hongqiao DING Xiaosong LI Xian International Business School Beijing Foreign Studies University Institute of Applied Mathematics Academy of Mathematics and Systems Science Chinese Academy of Sciences DECIDE Research Group Department of Computer and Systems Sciences Stockholm University

Along with the rapid development of economics and enhancement of industrialization, the power demand keeps rising and frequently creates mismatch between demand and supply in *** provides miscellaneous energy buy-back programs with great opportunities. Such programs, when activated, offer certain amount of financial compensations to participants for reducing their energy consumption during peak time. They aim at encouraging participants to shift their electricity usage from peak to non-peak time, and thereby release the demand pressure during peak time. This paper considers a periodic-review joint pricing and inventory decision model under an energy buy-back program over finite planning horizons, in which the compensation levels, setup cost and additive random demand function are incorporated. The objective is to maximize a manufacturer's expected total *** using Veinott's conditions, it is shown that the manufacturer's optimal decision is a state dependent(s, S, P) policy under a peak market condition, or partly an(s, S, A, P) policy under the normal market condition.

关键词： Additive demand function energy buy-back program joint pricing and inventory control Veinott＇s conditions.

来源：评论

学校读者我要写书评

暂无评论

Automatic neural robot controller design using Evolutionary Acquisition of Neural Topologies

Automatic neural robot controller design using Evolutionary ...

引用

19. Fachgesprach Autonome Mobile Systeme, AMS 2005 - 19th Expert Discussion on Autonomous Mobile systems, AMS 2005

作者： Kassahun, Yohannes Sommer, Gerald Department of Cognitive Systems Institute of Computer Science and Applied Mathematics Christian Albrechts University Olshausenstr. 40 D-24098 Kiel Germany

ISBN: (纸本)3540302913

In this paper we present an automatic design of neural controllers for robots using a method called Evolutionary Acquisition of Neural Topologies (EANT). The method evolves both the structure and weights of neural networks. It starts with networks of minimal structures determined by the domain expert and increases their complexity along the evolution path. It introduces an efficient and compact genetic encoding of neural networks onto a linear genome that enables one to evaluate the network without decoding it. The method uses a meta-level evolutionary process where new structures are explored at larger time-scale and existing structures are exploited at smaller time-scale. We demonstrate the method by designing a neural controller for a real robot whichshould be able to move continously in a given environment cluttered with obstacles. We first give an introduction to the evolutionary method and then describe the experiments and results obtained.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Structure multivector for local analysis of images 10th

Structure multivector for local analysis of images

引用

10th International Workshop on Theoretical Foundations of computer Vision, 2000

作者： Felsberg, Michael Sommer, Gerald Christian-Albrechts-University of Kiel Institute of Computer Science and Applied Mathematics Cognitive Systems Preußerstraße 1-9 Kiel24105 Germany

ISBN: (纸本)354042122X

The structure multivector is a new approach for analyzing the local properties of a two-dimensional signal (e.g. image). It combines the classical concepts of the structure tensor and the analytic signal in a new way. This has been made possible using a representation in the algebra of quaternions. The resulting method is linear and of low complexity. The filter-response includes local phase, local amplitude and local orientation of intrinsically one-dimensional neighborhoods in the signal. As for the structure tensor, the structure multivector field can be used to apply special filters to it for detecting features in images. © Springer-Verlag Berlin Heidelberg 2001.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Gabor wavelet networks for object representation 10th

Gabor wavelet networks for object representation

引用

10th International Workshop on Theoretical Foundations of computer Vision, 2000

作者： Krüger, Volker Sommer, Gerald Christian-Albrechts-University of Kiel Institute of Computer Science and Applied Mathematics Cognitive Systems Preußerstraße 1-9 Kiel24105 Germany

ISBN: (纸本)354042122X

In this article we want to introduce first the Gabor wavelet network as a model based approach for an effective and efficient object representation. The Gabor wavelet network has several advantages such as invariance to some degree with respect to translation, rotation and dilation. Furthermore, the use of Gabor filters ensured that geometrical and textural object features are encoded. The feasibility of the Gabor filters as a model for local object features ensures a considerable data reduction while at the same time allowing any desired precision of the object representation ranging from a sparse to a photo-realistic representation. In the second part of the paper we will present an approach for the estimation of a head pose that is based on the Gabor wavelet networks. © Springer-Verlag Berlin Heidelberg 2001.

关键词： Gabor filters

来源：评论

学校读者我要写书评

暂无评论

04491 - The Kiel Esterel Processor The Kiel Esterel Processor - A Semi-Custom, Configurable Reactive Processor - Dagstuhl Seminar -

04491 - The Kiel Esterel Processor The Kiel Esterel Processo...

引用

Synchronous Programming - SYNCHRON 2004

作者： Li, Xin von Hanxleden, Reinhard Real-Time and Embedded Systems Group Institute of Computer Science and Applied Mathematics Faculty of Engineering Christian-Albrechts-Universität zu Kiel Germany

The synchronous language Esterel is an established language for developing reactive systems. It gives an abstract, well-defined and executable description of the application, and can be synthesized into hardware and software. Typically, an Esterel program is first translated into other, lower-level languages (such as VHDL or C), and then compiled further. However, there is also the alternative of executing Esterel-like instructions directly. For example, in the REFLIX and RePIC projects, Roop et al. have augmented traditional processors with custom hardware to execute Esterel instructions. This patch strategy is a convenient approach, but has some shortages. We present the Kiel Esterel Processor (KEP), a semi-custom, configurable reactive processor for the direct execution of Esterel programs. It consists of a reactive core and scalable peripheral elements. KEP supports standard Esterel statements directly, except (so far) for the concurrency operator. Valued signals and counter functions in Esterel statements are supported by KEP. Due to its control path and its cooperation with elements, KEP obeys exact Esterel (preemption and priority) rules, including for example abort/weak abort (nests). © 2005 Dagstuhl Seminar Proceedings. All rights reserved.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

Static analysis of worst-case stack cache behavior 13

Static analysis of worst-case stack cache behavior

引用

21st International Conference on Real-Time Networks and systems, RTNS 2013

作者： Jordan, Alexander Brandner, Florian Schoeberl, Martin Compilers and Languages Group Institute of Computer Languages Vienna University of Technology Austria Embedded Systems Engineering Section Dep. of Applied Mathematics and Computer Science Technical University of Denmark Denmark

ISBN: (纸本)9781450320580

Utilizing a stack cache in a real-time system can aid predictability by avoiding interference that heap memory traffic causes on the data cache. While loads and stores are guaranteed cache hits, explicit operations are responsible for managing the stack cache. The behavior of these operations can be analyzed statically. We present algorithms that derive worst-case bounds on the latency-inducing operations of the stack cache. Their results can be used by a static WCET tool. By breaking the analysis down into subproblems that solve intra-procedural data-flow analysis and path searches on the call-graph, the worst-case bounds can be efficiently yet precisely determined. Our evaluation using the MiBench benchmark suite shows that only 37% and 21% of potential stack cache operations actually store to and load from memory, respectively. Analysis times are modest, on average running between 0.46s and 1.30s per benchmark, depending on the size of the stack cache. © 2013 ACM.

关键词： Real time systems

来源：评论

学校读者我要写书评

暂无评论

Reverse engineering the control law for schooling in zebrafish using virtual reality

引用

science Robotics 2025年第101期10卷 eadq6784页

作者： Li, Liang Nagy, Máté Amichay, Guy Wu, Ruiheng Wang, Wei Deussen, Oliver Rus, Daniela Couzin, Iain D. Department of Collective Behaviour Max Planck Institute of Animal Behavior Konstanz78464 Germany Centre for the Advanced Study of Collective Behaviour University of Konstanz Konstanz78464 Germany Department of Biology University of Konstanz Konstanz78464 Germany Department of Computer and Information Science University of Konstanz Konstanz78464 Germany MTA-ELTE "Lendület" Collective Behaviour Research Group Hungarian Academy of Sciences Budapest1117 Hungary Department of Biological Physics Eötvös Loránd University Budapest1117 Hungary Department of Engineering Sciences and Applied Mathematics Northwestern University EvanstonIL60208 United States Northwestern Institute on Complex Systems Northwestern University EvanstonIL60208 United States National Institute for Theory and Mathematics in Biology Northwestern University EvanstonIL60208 United States Department of Mechanical Engineering University of Wisconsin–Madison MadisonWI53706 United States Massachusetts Institute of Technology CambridgeMA02139 United States

Revealing the evolved mechanisms that give rise to collective behavior is a central objective in the study of cellular and organismal systems. In addition, understanding the algorithmic basis of social interactions in a causal and quantitative way offers an important foundation for subsequently quantifying social deficits. Here, with virtual reality technology, we used virtual robot fish to reverse engineer the sensory-motor control of social response during schooling in a vertebrate model: juvenile zebrafish (Danio rerio). In addition to providing a highly controlled means to understand how zebrafish translate visual input into movement decisions, networking our systems allowed real fish to swim and interact together in the same virtual world. Thus, we were able to directly test models of social interactions in situ. A key feature of social response is shown to be single- and multitarget-oriented pursuit. This is based on an egocentric representation of the positional information of conspecifics and is highly robust to incomplete sensory input. We demonstrated, including with a Turing test and a scalability test for pursuit behavior, that all key features of this behavior are accounted for by individuals following a simple experimentally derived proportional derivative control law, which we termed "BioPD." Because target pursuit is key to effective control of autonomous vehicles, we evaluated—as a proof of principle—the potential use of this simple evolved control law for human-engineered systems. In doing so, we found close-to-optimal pursuit performance in autonomous vehicle (terrestrial, airborne, and watercraft) pursuit while requiring limited system-specific tuning or optimization. Copyright © 2025 The Authors, some rights reserved

关键词： Invariance

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：