检索结果-内蒙古大学图书馆

BMEAUT at SemEval-2020 Task 2: Lexical entailment with semantic graphs 14

学校读者我要写书评

暂无评论

BMEAUT at SemEval-2020 Task 2: Lexical entailment with seman...

14th International Workshops on Semantic Evaluation, SemEval 2020

作者： Kovács, Ádám Gémes, Kinga Kornai, Andras Recski, Gábor Dept. of Automation and Applied Informatics Budapest University of Technology and Economics Hungary SZTAKI Institute of Computer Science TU Wien Austria

ISBN: (纸本)9781952148316

In this paper we present a novel rule-based, language independent method for determining lexical entailment relations using semantic representations built from Wiktionary definitions. Combined with a simple WordNet-based method our system achieves top scores on the English and Italian datasets of the Semeval-2020 task "Predicting Multilingual and Cross-lingual (graded) Lexical Entailment" (Glavaš et al., 2020). A detailed error analysis of our output uncovers future directions for improving both the semantic parsing method and the inference process on semantic graphs. reported in this document. © 2020 14th International Workshops on Semantic Evaluation, SemEval 2020 - co-located 28th International Conference on Computational Linguistics, COLING 2020, Proceedings. All rights reserved.

关键词： Semantics

K2-Hamiltonian Graphs: II

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Goedgebeur, Jan Renders, Jarne Wiener, Gábor Zamfirescu, Carol T. Department of Computer Science KU Leuven Campus Kulak-Kortrijk Kortrijk8500 Belgium Department of Applied Mathematics Computer Science and Statistics Ghent University Ghent9000 Belgium Department of Computer Science and Information Theory Faculty of Electrical Engineering and Informatics Budapest University of Technology and Economics Budapest Hungary Department of Mathematics Babeş-Bolyai University Cluj-Napoca Romania

In this paper we use theoretical and computational tools to continue our investigation of K2-hamiltonian graphs, that is, graphs in which the removal of any pair of adjacent vertices yields a hamiltonian graph, and their interplay with K1-hamiltonian graphs, that is, graphs in which every vertex-deleted subgraph is hamiltonian. Perhaps surprisingly, there exist graphs that are both K1- and K2-hamiltonian, yet non-hamiltonian, for example, the Petersen graph. Grünbaum conjectured that every planar K1-hamiltonian graph must itself be hamiltonian;Thomassen disproved this conjecture. Here we show that even planar graphs that are both K1- and K2-hamiltonian need not be hamiltonian, and that the number of such graphs grows at least exponentially. Motivated by results of Aldred, McKay, and Wormald, we determine for every integer n that is not 14 or 17 whether there exists a K2-hypohamiltonian, that is, non-hamiltonian and K2-hamiltonian, graph of order n, and characterise all orders for which such cubic graphs and such snarks exist. We also describe the smallest cubic planar graph which is K2-hypohamiltonian, as well as the smallest planar K2-hypohamiltonian graph of girth 5. We conclude with open problems and by correcting two inaccuracies from the first *** Codes 05C45, 05C38, 05C10, 05C85, 05C76 Copyright © 2023, The Authors. All rights reserved.

关键词： Hamiltonians

An Erasmus+ project: Interactive Course for Control Theory (ICCT) 8

学校读者我要写书评

暂无评论

An Erasmus+ project: Interactive Course for Control Theory (...

47th Annual Conference of the IEEE Industrial Electronics Society, IECON 2021

作者： Podrzaj, Primoz Finzgar, Miha Pirnar, Zan Ljubic, Sandi Lanzetta, Michele Pollini, Lorenzo Venturini, Pietro Pizag, Bertalan Urbin, Agnes Korondi, Peter Graff, Jozsef Budai, Csaba University of Ljubljana Laboratory for Mechatronics Production Systems and Automation Faculty of Mechanical Engineering Ljubljana Slovenia University of Rijeka Faculty of Engineering Department of Computer Engineering Rijeka Croatia University of Pisa Department of Industrial Engineering Department of Information Engineering Pisa Italy Budapest University of Technology and Economics Faculty of Mechanical Engineering Department of Mechatronics Optics and Mechanical Engineering Informatics Budapest Hungary

ISBN: (纸本)9781665435543

The paper presents an overview and the main results of the Erasmus+ project titled Interactive Course for Control Theory-ICCT. The project was started by four European Universities when they identified by a survey that students find Control Theory more challenging to understand compared to other subjects. The project's main goal was to develop a platform, which would offer examples from the field of Control Theory in an interactive way-ICCT. Frequently asked questions (FAQ) and test exam questions were added as well. To support the course, an accompanying textbook was prepared. At this point, the course and the accompanying textbook are available in English and the languages of the partner universities;since both of them are publicly available, we assume they will be offered in other languages in future. The benefits of implementing our interactive course are already visible in better survey results and improved grades in Control Theory-related subjects. © 2021 IEEE.

关键词： Python

Evaluating contextualized language models for Hungarian

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Ács, Judit Lévai, Dániel Nemeskey, Dávid Márk Kornai, András Department of Automation and Applied Informatics Budapest University of Technology and Economics Hungary Institute for Computer Science and Control Hungary Alfréd Rényi Institute of Mathematics Hungary

We present an extended comparison of contextualized language models for Hungarian. We compare huBERT, a Hungarian model against 4 multilingual models including the multilingual BERT model. We evaluate these models through three tasks, morphological probing, POS tagging and NER. We find that huBERT works better than the other models, often by a large margin, particularly near the global optimum (typically at the middle layers). We also find that huBERT tends to generate fewer subwords for one word and that using the last subword for token-level tasks is generally a better choice than using the first one. Copyright © 2021, The Authors. All rights reserved.

关键词： Computational linguistics

The Gutenberg Dialogue Dataset∗

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Csaky, Richard Recski, Gabor Department of Automation and Applied Informatics Budapest University of Technology and Economics Apollo.AI

Large datasets are essential for many NLP tasks. Current publicly available open-domain dialogue datasets offer a trade-off between size and quality (e.g. DailyDialog (Li et al., 2017b) vs. Opensubtitles1 (Tiedemann, 2012)). We aim to close this gap by building a high-quality dataset consisting of 14.8M utterances in English. We extract and process dialogues from publicly available online books. We present a detailed description of our pipeline and heuristics and an error analysis of extracted dialogues. Better response quality can be achieved in zero-shot and finetuning settings by training on our data than on the larger but much noisier Opensubtitles dataset. Researchers can easily build their versions of the dataset by adjusting various trade-off parameters. The code can be extended to further languages with limited effort2. Copyright © 2020, The Authors. All rights reserved.

关键词： Large dataset

Rolling Balance Board Robust Stabilization: A MID-based Design

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2022年第36期55卷 7-12页

作者： Amina Benarab Csenge A. Molnar Islam Boussaada Karim Trabelsi Tamas Insperger Silviu-Iulian Niculescu Université Paris-Saclay CNRS CentraleSupélec Inria Laboratoire des signaux et systèmes (L2S) 91190 Gif-sur-Yvette France Institut Polytechnique des Sciences Avancées (IPSA) 63 boulevard de Brandebourg 94200 Ivry-sur-Seine France Budapest University of Technology and Economics Faculty of Mechanical Engineering Department of Mechatronics Optics and Mechanical Engineering Informatics Budapest University of Technology and Economics Faculty of Mechanical Engineering Department of Applied Mechanics

This paper considers the stabilization of a rolling balance board by means of the multiplicity-induced-dominancy property. A two degree-of-freedom mechanical model of a human balancing on a rolling balance board is analyzed in the sagittal plane. The human body is modeled by an inverted pendulum which connects to the balance board through the ankle joint. The system is stabilized by the ankle torque managed by the central nervous system (CNS). The action of the CNS is modeled by a delayed full state feedback: a pointwise delay stands for all latencies in the neuromechanical system (reaction time, neuromechanical lag, etc.). The aim of the paper is to achieve a good occurrence in terms of the decay rate, it shows the links with dominancy and with the exponential stability property of the solution.

关键词： Human balance exponential stability Multiplicity-Induced-Dominancy

Color-avoiding percolation in edge-colored Erdős–Rényi graphs

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ráth, Balázs Varga, Kitti Fekete, Panna Tímea Molontay, Roland Afréd Rényi Institute of Mathematics Hungary Department of Stochastics Institute of Mathematics Budapest University of Technology and Economics Hungary ELKH-BME Stochastics Research Group Chile Department of Computer Science and Information Theory Faculty of Electrical Engineering and Informatics Budapest University of Technology and Economics Hungary ELKH-ELTE Egerváry Research Group Hungary Department of Applied Analysis and Computational Mathematics Eötvös Loránd University Hungary

We study a variant of the color-avoiding percolation model introduced by Krause et al., namely we investigate the color-avoiding bond percolation setup on (not necessarily properly) edge-colored Erd̋ os–Rényi random graphs. We say that two vertices are color-avoiding connected in an edge-colored graph if after the removal of the edges of any color, they are in the same component in the remaining graph. The color-avoiding connected components of an edge-colored graph are maximal sets of vertices such that any two of them are color-avoiding connected. We consider the fraction of vertices contained in color-avoiding connected components of a given size as well as the fraction of vertices contained in the giant color-avoiding connected component. Under some mild assumptions on the color-densities, we prove that these quantities converge and the limits can be expressed in terms of probabilities associated to edge-colored branching process trees. We provide explicit formulas for the limit of the normalized size of the giant color-avoiding connected component, and in the two-colored case we also provide explicit formulas for the limit of the fraction of vertices contained in color-avoiding connected components of a given size. Copyright © 2022, The Authors. All rights reserved.

关键词： Color

Cross-period single phase shift control technique for high power and low frequency dual active bridge converters 20

学校读者我要写书评

暂无评论

Cross-period single phase shift control technique for high p...

20th International Conference on Electrical Drives and Power Electronics, EDPE 2019 - 9th Joint Slovakian-Croatian Conference

作者： Vereb, Szabolcs Futo, Andras Suto, Zoltan Blogh, Attila Varjasi, Istvan Budapest University of Technology and Economics Department of Automation and Applied Informatics Budapest Hungary

ISBN: (纸本)9781728103907;9781728103891

Nowadays, the need for isolated bidirectional DC-to-DC converters is increasing, as electric energy storage is gaining more popularity. For high power applications, the Dual Active Bridge (DAB) DC-to-DC converters are widely used. Various control methods were proposed throughout the years, but none of them is practical to use with low frequency and low leakage transformers. In this paper, a novel control algorithm is presented, which is specifically designed for these applications and has outstanding transient response characteristics. The behavior is analyzed and compared to Single Phase Shift control thorough hardware-in-the-loop simulation. © 2019 IEEE.

关键词： 400Hz transformer DAB Dual Active Bridge Isolated converter

The role of interpretable patterns in deep learning for morphology

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ács, Judit Kornai, András Department of Automation and Applied Informatics Budapest University of Technology and Economics Hungary Institute for Computer Science and Control Hungary

We examine the role of character patterns in three tasks: morphological analysis, lemmatization and copy. We use a modified version of the standard sequence-to-sequence model, where the encoder is a pattern matching network. Each pattern scores all possible N character long subwords (substrings) on the source side, and the highest scoring subword’s score is used to initialize the decoder as well as the input to the attention mechanism. This method allows learning which subwords of the input are important for generating the output. By training the models on the same source but different target, we can compare what subwords are important for different tasks and how they relate to each other. We define a similarity metric, a generalized form of the Jaccard similarity, and assign a similarity score to each pair of the three tasks that work on the same source but may differ in target. We examine how these three tasks are related to each other in 12 languages. Our code is publicly available. © 2020, CC BY-SA.

关键词： Pattern matching