检索结果-内蒙古大学图书馆

6th Conference on Cloud and Internet of Things (CIoT)

作者： Zhang Weijia Jayakody, Dushantha Nalin K. Tomsk Polytech Univ Sch Comp Sci & Robot Tomsk Russia Lusofona Univ COPELABS P-1700097 Lisbon Portugal

ISBN: (纸本)9798350396690

The belief propagation (BP) algorithm for low density parity check (LDPC) codes has proven to have excellent performance and is close to the Shannon limit. However, the high computational complexity is not conducive to realistic use. Although the min sum algorithm (MSA) improves the computational convenience and applicability. It also weakens its decoding capability. In order to improve the decoding performance of MSA and exploit the potential, this paper propose an adaptive exponential correction factor. This can improve the decoding performance of MSA. Adaptive exponential min sum algorithm (AEMSA) demonstrate a noticeable improvement over the normalized min sum algorithm (NMSA) and the BP algorithm in high signal to noise ratio (SNR). AEMSA is also a simple, efficient, low power consuming and highly potential decoding algorithm that can be applied in hardware implementations.

关键词： LDPC Codes decoding algorithms Min Sum Adaptive Exponential Internet of Things (IoTs)

来源：评论

学校读者我要写书评

暂无评论

UnsupervisedWord Segmentation with Bi-directional Neural Language Model

引用

ACM TRANSACTIONS ON ASIAN AND LOW-RESOURCE LANGUAGE INFORMATION PROCESSING 2023年第1期22卷 1-16页

作者： Wang, Lihao Zheng, Xiaoqing Fudan Univ 220 Handan Rd Shanghai Peoples R China

We propose an unsupervised word segmentation model, in which for each unlabelled sentence sample, the learning objective is to maximize the generation probability of the sentence given its all possible segmentations. Such a generation probability can be factorized into the likelihood of each possible segment given the context in a recursive way. To capture both the long- and short-term dependencies, we propose to use a bi-directional neural language model to better extract the features of the segment's context. Two decoding algorithms were also developed to combine the context features from both directions to generate the final segmentation at the inference time, which helps to reconcile word-boundary ambiguities. Experimental results show that our context-sensitive unsupervised segmentation model achieved state-of-the-art at different evaluation settings on various datasets for Chinese, and the comparable result for Thai.

关键词： Unsupervised word segmentation bi-directional neural language model recurrent neural networks context-sensitive segmentation decoding algorithms

来源：评论

学校读者我要写书评

暂无评论

E2E Segmenter: Joint Segmenting and decoding for Long-Form ASR 23

E2E Segmenter: Joint Segmenting and Decoding for Long-Form A...

引用

Interspeech Conference

作者： Huang, W. Ronny Chang, Shuo-yiin Rybach, David Prabhavalkar, Rohit Sainath, Tara N. Allauzen, Cyril Peyser, Cal Lu, Zhiyun Google Res Mountain View CA 94043 USA

Improving the performance of end-to-end ASR models on long utterances ranging from minutes to hours is an ongoing challenge in speech recognition. A common solution is to segment the audio in advance using a separate voice activity detector (VAD) that decides segment boundary locations based purely on acoustic speech/non-speech information. VAD segmenters, however, may be sub-optimal for real-world speech where, e.g., a complete sentence that should be taken as a whole may contain hesitations in the middle ("set an alarm for... 5 o'clock"). We propose to replace the VAD with an end-to-end ASR model capable of predicting segment boundaries in a streaming fashion, allowing the segmentation decision to be conditioned not only on better acoustic features but also on semantic features from the decoded text with negligible extra computation. In experiments on real world long-form audio (YouTube) with lengths of up to 30 minutes long, we demonstrate 8.5% relative WER improvement and 250 ms reduction in median end-of-segment latency compared to the VAD baseline on a state-of-the-art Conformer RNN-T model.

关键词： speech recognition speech segmentation decoding algorithms beam search

来源：评论

学校读者我要写书评

暂无评论

A matrix based list decoding algorithm for linear codes over integer residue rings

引用

LINEAR ALGEBRA AND ITS APPLICATIONS 2021年 614卷 376-393页

作者： Napp, Diego Pinto, Raquel Sacikara, Elif Toste, Marisa Univ Alicante Dept Math Alicante Spain Univ Zurich Inst Math Zurich Switzerland Univ Aveiro Dept Math CIDMA Ctr Res & Dev Math & Applicat Aveiro Portugal CIDMA Ctr Res & Dev Math & Applicat Aveiro Portugal Polytech Inst Coimbra Super Sch Technol & Management Olveira Hosp Coimbra Portugal

In this paper we address the problem of list decoding of linear codes over an integer residue ring Z(q), where q is a power of a prime p. The proposed procedure exploits a particular matrix representation of the linear code over Z(pr), called the standard form, and the p-adic expansion of the to-be-decoded vector. In particular, we focus on the erasure channel in which the location of the errors is known. This problem then boils down to solving a system of linear equations with coefficients in Z(pr). From the parity-check matrix representations of the code we recursively select certain equations that a codeword must satisfy and have coefficients only in the field p(r-1)Z(pr),This yields a step by step procedure obtaining a list of the closest codewords to a given received vector with some of its coordinates erased. We show that such an algorithm actually computes all possible erased coordinates, that is, the provided list is minimal. (C) 2020 Published by Elsevier Inc.

关键词： Finite rings Linear codes over finite rings Erasure channel decoding algorithms Matrix representations Parity-check matrix

来源：评论

学校读者我要写书评

暂无评论

Near-Optimal Sparsity-Constrained Group Testing: Improved Bounds and algorithms

引用

IEEE TRANSACTIONS ON INFORMATION THEORY 2022年第5期68卷 3253-3280页

作者： Gebhard, Oliver Hahn-Klimroth, Max Parczyk, Olaf Penschuck, Manuel Rolvien, Maurice Scarlett, Jonathan Tan, Nelvin Tech Univ Dortmund Fac Comp Sci D-44227 Dortmund Germany FU Berlin Dept Math & Comp Sci D-14195 Berlin Germany Goethe Univ Frankfurt Inst Comp Sci D-60325 Frankfurt Germany Natl Univ Singapore Dept Comp Sci Singapore 117418 Singapore Univ Cambridge Dept Engn Cambridge CB2 1PZ England

Recent advances in noiseless non-adaptive group testing have led to a precise asymptotic characterization of the number of tests required for high-probability recovery in the sublinear regime k = n(theta) (with theta is an element of(0, 1)), with n individuals among which k are infected. However, the required number of tests may increase substantially under real-world practical constraints, notably including bounds on the maximum number Delta of tests an individual can be placed in, or the maximum number Gamma of individuals in a given test. While previous works have given recovery guarantees for these settings, significant gaps remain between the achievability and converse bounds. In this paper, we substantially or completely close several of the most prominent gaps. In the case of Delta-divisible items, we show that the definite defectives (DD) algorithm coupled with a random regular design is asymptotically optimal in dense scaling regimes, and optimal to within a factor of e more generally;we establish this by strengthening both the best known achievability and converse bounds. In the case of Gamma-sized tests, we provide a comprehensive analysis of the regime Gamma = Theta(1), and again establish a precise threshold proving the asymptotic optimality of SCOMP (a slight refinement of DD) equipped with a tailored pooling scheme. Finally, for each of these two settings, we provide near-optimal adaptive algorithms based on sequential splitting, and provably demonstrate gaps between the performance of optimal adaptive and non-adaptive algorithms.

关键词： Group testing sparsity information-theoretic limits decoding algorithms bayesian inference fragile complexity

来源：评论

学校读者我要写书评

暂无评论

Review of brain encoding and decoding mechanisms for EEG-based brain-computer interface

引用

COGNITIVE NEURODYNAMICS 2021年第4期15卷 569-584页

作者： Xu, Lichao Xu, Minpeng Jung, Tzzy-Ping Ming, Dong Tianjin Univ Acad Med Engn & Translat Med Tianjin Peoples R China Tianjin Univ Coll Precis Instruments & Optoelect Engn Dept Biomed Engn Tianjin Peoples R China Univ Calif San Diego Swartz Ctr Computat Neurosci San Diego CA 92103 USA

A brain-computer interface (BCI) can connect humans and machines directly and has achieved successful applications in the past few decades. Many new BCI paradigms and algorithms have been developed in recent years. Therefore, it is necessary to review new progress in BCIs. This paper summarizes progress for EEG-based BCIs from the perspective of encoding paradigms and decoding algorithms, which are two key elements of BCI systems. Encoding paradigms are grouped by their underlying neural meachanisms, namely sensory- and motor-related, vision-related, cognition-related and hybrid paradigms. decoding algorithms are reviewed in four categories, namely decomposition algorithms, Riemannian geometry, deep learning and transfer learning. This review will provide a comprehensive overview of both modern primary paradigms and algorithms, making it helpful for those who are developing BCI systems.

关键词： EEG BCI Encoding paradigms decoding algorithms Review

来源：评论

学校读者我要写书评

暂无评论

E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model 48

E2E Segmentation in a Two-Pass Cascaded Encoder ASR Model

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Huang, W. Ronny Chang, Shuo-Yiin Sainath, Tara N. He, Yanzhang Rybach, David David, Robert Prabhavalkar, Rohit Allauzen, Cyril Peyser, Cal Strohman, Trevor D. Google Research United States

ISBN: (纸本)9781728163277

We explore unifying a neural segmenter with two-pass cascaded encoder ASR into a single model. A key challenge is allowing the segmenter (which runs in real-time, synchronously with the decoder) to finalize the non-causal 2nd pass (which runs 900 ms behind real-time) without introducing user-perceived latency or deletion errors during inference. We propose a design where the neural segmenter is integrated with the causal 1st pass decoder to emit a end-of-segment (EOS) signal in real-time. The EOS signal is then used to finalize the non-causal 2nd pass. We experiment with different ways to finalize the 2nd pass, and find that a dummy frame injection strategy allows for simultaneous high quality 2nd pass results and low finalization latency. On a real-world long-form captioning task (YouTube), we achieve 2.4% relative WER and 140 ms EOS latency gains over a baseline VAD-based segmenter with the same cascaded encoder. © 2023 IEEE.

关键词： ASR decoding algorithms segmentation

来源：评论

学校读者我要写书评

暂无评论

Reed-Muller Codes: Theory and algorithms

引用

IEEE TRANSACTIONS ON INFORMATION THEORY 2021年第6期67卷 3251-3277页

作者： Abbe, Emmanuel Shpilka, Amir Ye, Min Ecole Polytech Fed Lausanne Math Inst CH-1015 Lausanne Switzerland Ecole Polytech Fed Lausanne Sch Comp & Commun Sci CH-1015 Lausanne Switzerland Princeton Univ Program Appl & Computat Math Princeton NJ 08544 USA Princeton Univ Dept Elect Engn Princeton NJ 08544 USA Tel Aviv Univ Blavatnik Sch Comp Sci IL-60198 Tel Aviv Israel Tsinghua Shenzhen Int Grad Sch Tsinghua Berkeley Shenzhen Inst Data Sci & Informat Technol Res Ctr Shenzhen 518055 Peoples R China

Reed-Muller (RM) codes are among the oldest, simplest and perhaps most ubiquitous family of codes. They are used in many areas of coding theory in both electrical engineering and computer science. Yet, many of their important properties are still under investigation. This paper covers some of the recent developments regarding the weight enumerator and the capacity-achieving properties of RM codes, as well as some of the algorithmic developments. In particular, the paper discusses the recent connections established between RM codes, thresholds of Boolean functions, polarization theory, hypercontractivity, and the techniques of approximating low weight codewords using lower degree polynomials (when codewords are viewed as evaluation vectors of degree r polynomials in m variables). It then overviews some of the algorithms for decoding RM codes. It covers both algorithms with provable performance guarantees for every block length, as well as algorithms with state-of-the-art performances in practical regimes, which do not perform as well for large block length. Finally, the paper concludes with a few open problems.

关键词： Reed-Muller codes Shannon capacity weight enumerator decoding algorithms polarization

来源：评论

学校读者我要写书评

暂无评论

Correcting Spanning Errors With a Fractal Code

引用

IEEE TRANSACTIONS ON INFORMATION THEORY 2021年第7期67卷 4504-4516页

作者： Nixon, Georgia M. Brown, Benjamin J. Univ Sydney Sch Phys Ctr Engn Quantum Syst Sydney NSW 2006 Australia Univ Cambridge Cavendish Lab Cambridge CB3 0HE England Univ Sydney Sch Phys Ctr Engn Quantum Syst Sydney NSW 2006 Australia

The strongly correlated systems we use to realise quantum error-correcting codes may give rise to high-weight, problematic errors. Encouragingly, we can expect local quantum error-correcting codes with no string-like logical operators - such as the cubic code - to be robust to highly correlated, one-dimensional errors that span their lattice. The challenge remains to design decoding algorithms that utilise the high distance of these codes. Here, we begin the development of such algorithms by proposing an efficient decoder for the 'Fibonacci code';a two-dimensional classical code that mimics the fractal nature of the cubic code. Our iterative decoder finds a correction through repeated use of minimum-weight perfect matching by exploiting symmetries of the code. We perform numerical experiments that show our decoder is robust to one-dimensional, correlated errors. First, using a bit-flip noise model at low error rates, we find that our decoder demonstrates a logical failure rate that scales super exponentially in the linear size of the lattice. In contrast, a decoder that could not tolerate spanning errors would not achieve this rapid decay in failure rate with increasing system size. We also find a finite threshold using a spanning noise model that introduces string-like errors that stretch along full rows and columns of the lattice. These results provide direct evidence that our decoder is robust to one-dimensional, correlated errors that span the lattice.

关键词： decoding Lattices Fractals Faces Qubit Runtime Error correction codes Classical coding theory fractal codes decoding algorithms minimum-weight perfect matching quantum error correction fracton topological codes

来源：评论

学校读者我要写书评

暂无评论

A survival duration-guided NSGA-III for sustainable flexible job shop scheduling problem considering dual resources

IET COLLABORATIVE INTELLIGENT MANUFACTURING

引用

IET COLLABORATIVE INTELLIGENT MANUFACTURING 2021年第2期3卷 119-130页

作者： Li Hongyu Wu Xiuli Univ Sci & Technol Beijing Sch Mech Engn Dept Logist Engn Beijing Peoples R China

Considering the increasing concern on sustainable development from manufacturers, focus is given to three kinds of indicators of sustainable development, that is, economy, environment, and society, and schedule two types of resource, that is machines and workers, simultaneously in the classical flexible job shop scheduling problem. The authors define it as a sustainable flexible job shop scheduling problem considering dual resources (SFJSPCDR). First, a model of the SFJSPCDR is formulated to optimise the makespan, the energy consumption, and the ergonomic risk simultaneously. Second, an improved survival duration-guided NSGA-III algorithm (SDG-NSGA-III) is proposed to solve SFJSPCDR. The survival duration of each individual determines whether it takes part in generating offspring. In order to balance the energy consumption and ergonomic risk while minimising the makespan, a double-low decoding algorithm is proposed, which is composed of two decoding algorithms. Cross-generation selection is employed with the non-dominated sorting, and the next-generation population is selected according to the reference point-based selection strategy. In addition, a restart strategy is also integrated to improve the exploration and exploitation performance of the SDG-NSGA-III algorithm. Finally, a group of experiments are carried out and the results prove the effectiveness of the proposed algorithm.

关键词： decoding algorithms sustainable flexible job shop scheduling dual resources ergonomic risk sustainable development job shop scheduling ergonomics energy consumption survival duration-guided NSGA-III algorithm cross-generation selection minimisation SDG-NSGA-III algorithm nondominated sorting SFJSPCDR next-generation population genetic algorithms

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：