检索结果-内蒙古大学图书馆

Geometric Reinforcement Learning For Robotic Manipulation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Alhousani, Naseem Saveriano, Matteo Sevinc, Ibrahim Abdulkuddus, Talha Kose, Hatice Abu-Dakka, Fares J. Faculty of Computer and Informatics Engineering Istanbul Technical University Maslak Sarıyer Istanbul34467 Turkey University of Trento Trento38123 Italy ILITRON Enerji ve Bilgi Teknolojileri A.Ş Sultan Selim Mahallesi Akyol Sanayi Sitesi Çıkmazı No:10/1 Kagıthane Istanbul Turkey MCFLY Robot Teknolojileri A.Ş Huzur Mah. Ahmet Bayman Cad. No: 2a Sarıyer Istanbul Turkey Technical University of Munich Germany Germany

Reinforcement learning (RL) is a popular technique that allows an agent to learn by trial and error while interacting with a dynamic environment. The traditional Reinforcement Learning (RL) approach has been successful in learning and predicting Euclidean robotic manipulation skills such as positions, velocities, and forces. However, in robotics, it is common to encounter non-Euclidean data such as orientation or stiffness, and failing to account for their geometric nature can negatively impact learning accuracy and performance. In this paper, to address this challenge, we propose a novel framework for RL that leverages Riemannian geometry, which we call Geometric Reinforcement Learning (G-RL), to enable agents to learn robotic manipulation skills with non-Euclidean data. Specifically, G-RL utilizes the tangent space in two ways: a tangent space for parameterization and a local tangent space for mapping to a non-Euclidean manifold. The policy is learned in the parameterization tangent space, which remains constant throughout the training. The policy is then transferred to the local tangent space via parallel transport and projected onto the non-Euclidean manifold. The local tangent space changes over time to remain within the neighborhood of the current manifold point, reducing the approximation error. Therefore, by introducing a geometrically grounded pre- and post-processing step into the traditional RL pipeline, our G-RL framework enables several model-free algorithms designed for Euclidean space to learn from non-Euclidean data without modifications. Experimental results, obtained both in simulation and on a real robot, support our hypothesis that G-RL is more accurate and converges to a better solution than approximating non-Euclidean data. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

Gamified crowdsourcing for idiom corpora construction

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Eryigit, Gülşen Şentaş, Ali Monti, Johanna NLP Research Group Faculty of Computer&Informatics Istanbul Technical University Istanbul Turkey Department of Artificial Intelligence & Data Engineering Istanbul Technical University Istanbul Turkey UNIOR NLP Research Group Department of Literary Linguistic and Comparative Studies University of Naples L’Orientale Italy

Learning idiomatic expressions is seen as one of the most challenging stages in second language learning because of their unpredictable meaning. A similar situation holds for their identification within natural language processing applications such as machine translation and parsing. The lack of high-quality usage samples exacerbates this challenge not only for humans but also for artificial intelligence systems. This article introduces a gamified crowdsourcing approach for collecting language learning materials for idiomatic expressions;a messaging bot is designed as an asynchronous multiplayer game for native speakers who compete with each other while providing idiomatic and nonidiomatic usage examples and rating other players’ entries. As opposed to classical crowdprocessing annotation efforts in the field, for the first time in the literature, a crowdcreating & crowdrating approach is implemented and tested for idiom corpora construction. The approach is language independent and evaluated on two languages in comparison to traditional data preparation techniques in the field. The reaction of the crowd is monitored under different motivational means (namely, gamification affordances and monetary rewards). The results reveal that the proposed approach is powerful in collecting the targeted materials, and although being an explicit crowdsourcing approach, it is found entertaining and useful by the crowd. The approach has been shown to have the potential to speed up the construction of idiom corpora for different natural languages to be used as second language learning material, training data for supervised idiom identification systems, or samples for lexicographic studies. Copyright © 2021, The Authors. All rights reserved.

关键词： Crowdsourcing

Constructing a Cryptocurrency-Price Prediction Model Using Deep Learning

学校读者我要写书评

暂无评论

Constructing a Cryptocurrency-Price Prediction Model Using D...

International Conference on engineering and Emerging Technologies (ICEET)

作者： Toai Kim Tran Thanh Thi Tuyet Le Thinh Tien Bui Vang Quang Dang Roman Senkerik Faculty of Economics Ho Chi Minh City University of Technology and Education No 1 Vo Van Ngan Street Linh Chieu Ward Thu Duc City Vietnam Dept. of Computer Science Faculty of Electrical Engineering and Computer Science VSB-Technical University of Ostrava 17. Listopadu 2172/15 Ostrava-Poruba Czech Repulic Faculty of Applied Informatics Tomas Bata University in Zlin T. G. Masaryka 5555 Zlin Czech Republic

ISBN: (纸本)9781665491075

The purpose of this study is to discover the optimal Deep Learning model for Bitcoin prediction among the Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM). Our empirical results indicate that LSTM is the optimal model for predicting Bitcoin price and trend with the prediction accuracy of 88.9%. Our study serves as a stepping stone for novice cryptocurrency investors and future studies of more advanced and sophisticated algorithms. Finally, given that the ideal model for predicting the price of cryptocurrencies is still a topic of controversy, the findings of this study will serve as a valuable empirical resource for future studies.

关键词： Deep learning Recurrent neural networks Bitcoin Predictive models Prediction algorithms Market research Convolutional neural networks

A Computational Framework for Dissociating Development-Related from Individually Variable Flexibility in Regional Modularity Assignment in Early Infancy 23rd

学校读者我要写书评

暂无评论

A Computational Framework for Dissociating Development-Relat...

23rd International Conference on Medical Image Computing and computer-Assisted Intervention, MICCAI 2020

作者： Soussia, Mayssa Wen, Xuyun Zhou, Zhen Jin, Bing Kam, Tae-Eui Hsu, Li-Ming Wu, Zhengwang Li, Gang Wang, Li Rekik, Islem Lin, Weili Shen, Dinggang Zhang, Han University of North Carolina at Chapel Hill Chapel HillNC United States BASIRA Lab Faculty of Computer and Informatics Istanbul Technical University Istanbul Turkey Department of Biomedical Engineering University of North Carolina at Chapel Hill Chapel HillNC United States

ISBN: (纸本)9783030597276

Functional brain development in early infancy is a highly dynamic and complex process. Understanding each brain regions topological role and its development in the brain functional connectivity (FC) networks is essential for early disorder detection. A handful of previous studies have mostly focused on how FC network is changing regarding age. These approaches inevitably overlook the effect of individual variability for those at the same age that could shape unique cognitive capabilities and personalities among infants. With that in mind, we propose a novel computational framework based on across-subject across-age multilayer network analysis with a fully automatic (for parameter optimization), robust community detection algorithm. By detecting group consistent modules without losing individual information, this method allows a first-ever dissociation analysis of the two variability sources – age dependency and individual specificity – that greatly shape early brain development. This method is applied to a large cohort of 02 years old infants functional MRI data during natural sleep. We not only detected the brain regions with greatest flexibility in this early developmental period but also identified five categories of brain regions with distinct development-related and individually variable flexibility changes. Our method is highly valuable for more thorough understanding of the early brain functional organizations and sheds light on early developmental abnormality detection. © 2020, Springer Nature Switzerland AG.

关键词： Brain

Airplane type identification based on mask RCNN and drone images

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Alshaibani, Waleed Tawfik Abdullah Helvaci, Mustafa Shayea, Ibraheem Saad, Sawsan A. Azizan, Azizul Yakub, Fitri Institute of Informatics Satellite Communication and Remote Sensing Program Istanbul Technical University ITU Ayazaga Campus Institute of Informatics Building Sariyer Istanbul34469 Turkey Istanbul34469 Turkey Faculty of Engineering and Built Environment Universiti Sains Islam Malaysia Negeri Sembilan Bandar BaruNilai71800 Malaysia Department of Computer Engineering College of Computer Science and Engineering University of Ha'il Saudi Arabia Advance Informatics Department Razak Faculty of Technology and Informatics Universiti Teknologi Malaysia Kuala Lumpur54100 Malaysia

For dealing with traffic bottlenecks at airports, aircraft object detection is insufficient. Every airport generally has a variety of planes with various physical and technological requirements as well as diverse service requirements. Detecting the presence of new planes will not address all traffic congestion issues. Identifying the type of airplane, on the other hand, will entirely fix the problem because it will offer important information about the plane's technical specifications (i.e., the time it needs to be served and its appropriate place in the airport). Several studies have provided various contributions to address airport traffic jams;however, their ultimate goal was to determine the existence of airplane objects. This paper provides a practical approach to identify the type of airplane in airports depending on the results provided by the airplane detection process using mask region convolution neural network. The key feature employed to identify the type of airplane is the surface area calculated based on the results of airplane detection. The surface area is used to assess the estimated cabin length which is considered as an additional key feature for identifying the airplane type. The length of any detected plane may be calculated by measuring the distance between the detected plane's two furthest points. The suggested approach's performance is assessed using average accuracies and a confusion matrix. The findings show that this method is dependable. This method will greatly aid in the management of airport traffic congestion. Copyright © 2021, The Authors. All rights reserved.

关键词： Aircraft detection

Inter-domain alignment for predicting high-resolution brain networks using teacher-student learning

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Demir, Başar Bessadok, Alaa Rekik, Islem BASIRA Lab Faculty of Computer and Informatics Istanbul Technical University Istanbul Turkey Higher Institute of Informatics and Communication Technologies University of Sousse 4011 Tunisia National Engineering School of Sousse University of Sousse LATIS- Laboratory of Advanced Technology and Intelligent Systems Sousse4023 Tunisia

Accurate and automated super-resolution image synthesis is highly desired since it has the great potential to circumvent the need for acquiring high-cost medical scans and a time-consuming preprocessing pipeline of neuroimaging data. However, existing deep learning frameworks are solely designed to predict high-resolution (HR) image from a low-resolution (LR) one, which limits their generalization ability to brain graphs (i.e., connectomes). A small body of works has focused on superresolving brain graphs where the goal is to predict a HR graph from a single LR graph. Although promising, existing works mainly focus on superresolving graphs belonging to the same domain (e.g., functional), overlooking the domain fracture existing between multimodal brain data distributions (e.g., morphological and structural). To this aim, we propose a novel inter-domain adaptation framework namely, Learn to SuperResolve Brain Graphs with Knowledge Distillation Network (L2S-KDnet), which adopts a teacher-student paradigm to superresolve brain graphs. Our teacher network is a graph encoder-decoder that firstly learns the LR brain graph embeddings, and secondly learns how to align the resulting latent representations to the HR ground truth data distribution using an adversarial regularization. Ultimately, it decodes the HR graphs from the aligned embeddings. Next, our student network learns the knowledge of the aligned brain graphs as well as the topological structure of the predicted HR graphs transferred from the teacher. We further leverage the decoder of the teacher to optimize the student network. In such a way, we are not only bringing the learned embeddings from both networks closer to each other but also their predicted HR graphs. L2S-KDnet presents the first TS architecture tailored for brain graph super-resolution synthesis that is based on inter-domain alignment. Our experimental results demonstrate substantial performance gains over benchmark methods. Our code is available a

关键词： Graphic methods

A Variational Graph Autoencoder for Manipulation Action Recognition and Prediction

学校读者我要写书评

暂无评论

A Variational Graph Autoencoder for Manipulation Action Reco...

International Conference on Advanced Robotics (ICAR)

作者： Gamze Akyol Sanem Sariel Eren Erdal Aksoy Faculty of Computer and Informatics Engineering Artificial Intelligence and Robotics Laboratory Istanbul Technical University Maslak Turkey School of Information Technology Center for Applied Intelligent Systems Research Halmstad University Halmstad Sweden

ISBN: (纸本)9781665436854

Despite decades of research, understanding human manipulation activities is, and has always been, one of the most attractive and challenging research topics in computer vision and robotics. Recognition and prediction of observed human manipulation actions have their roots in the applications related to, for instance, human-robot interaction and robot learning from demonstration. The current research trend heavily relies on advanced convolutional neural networks to process the structured Euclidean data, such as RGB camera images. These networks, however, come with immense computational complexity to be able to process high dimensional raw data. Different from the related works, we here introduce a deep graph autoencoder to jointly learn recognition and prediction of manipulation tasks from symbolic scene graphs, instead of relying on the structured Euclidean data. Our network has a variational autoencoder structure with two branches: one for identifying the input graph type and one for predicting the future graphs. The input of the proposed network is a set of semantic graphs which store the spatial relations between subjects and objects in the scene. The network output is a label set representing the detected and predicted class types. We benchmark our new model against different state-of-the-art methods on two different datasets, MANIAC and MSRC-9, and show that our proposed model can achieve better performance. We also release our source code https://***/gamzeakyol/GNet.

关键词： computer vision Semantics Robot vision systems Human-robot interaction Predictive models Benchmark testing Market research

A variational graph autoencoder for manipulation action recognition and prediction

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Akyol, Gamze Sariel, Sanem Aksoy, Eren Erdal Artificial Intelligence and Robotics Laboratory Faculty of Computer and Informatics Engineering Istanbul Technical University Maslak Turkey School of Information Technology Center for Applied Intelligent Systems Research Halmstad University Halmstad Sweden

关键词： Forecasting

Organization of Regional Situational Centers of the Intelligent System «cONTROL-TEA» using UAVs 6

学校读者我要写书评

暂无评论

Organization of Regional Situational Centers of the Intellig...

6th IEEE International Conference on Actual Problems of Unmanned Aerial Vehicles Development, APUAVD 2021

作者： Melkumian, Kateryna Pisarenko, Julia Koval, Alexander Faculty of Informatics and Computer Engineering National Technical University of Ukraine Igor Sikorsky Kyiv Polytechnic Institute of Ukraine Kyiv Ukraine Lab. 236 of Virtual Environment Systems for Organization of Scientific Research V.M. Glushkov Institute of Cybernetics National Academy of Sciences of Ukraine Kyiv Ukraine V.M. Glushkov Institute of Cybernetics National Academy of Sciences of Ukraine Dept. #265 of Mathematical Problems of Applied Informatics Kyiv Ukraine

ISBN: (纸本)9781665438223

Some questions of creating the intellectual system «CONTROL-TEA» concept was considered. The issues of creating organizational support for the situational center of intellectual system «CONTROL-TEA» are reviewed. The structure of the information storage was developed. There is a need for their certification and registration in city services. In order to combine them into a single system, the smart city will be the task manager, and the DAVs will be their executor, it is proposed to organize regional situational centers that will coordinate the work. The regional situation center provides two-way communication between customers (public, municipal, private) and its contractors (UAVs), monitors traffic, provides a variety of operations and, along with the IOT, serves as a data provider. © 2021 IEEE.

关键词： Intelligent systems