Filters








157,074 Hits in 3.3 sec

Comparing coordination models and architectures using embeddings

Marcello M. Bonsangue, Joost N. Kok, Gianluigi Zavattaro
2003 Science of Computer Programming  
We will use this notion to present equivalence and di erence results for several coordination models based on components that communicate either through an unordered broadcast, through an atomic broadcast  ...  The new notion, called architectural embedding, is suitable for the comparison of di erent communication mechanisms, and gives rise to a natural notion of implementability.  ...  This provides a useful framework for the relative comparison of coordination languages, models, and architectures; for example, it is possible to compare architectures in isolation, i.e., independently  ... 
doi:10.1016/s0167-6423(02)00086-2 fatcat:qw7z54kpxbc5jnsl3ssosafzvm

Issues in Coordination Languages and Architectures

Paolo Ciancarini, Alexander Wolf
2003 Science of Computer Programming  
Acknowledgements We thank the anonymous reviewers who helped us in the selection: their help was invaluable in putting together this special issue.  ...  "Comparing coordination models and architectures using embeddings" by Bonsangue, Kok, and Zavattaro, is a paper comparing computational expressiveness of some basic coordination models using the technique  ...  "Erratic Fudgets: a semantic theory for an embedded coordination language" by Moran, Sands, and Carlsson, introduces a theoretical model and a functional language including some special coordination features  ... 
doi:10.1016/s0167-6423(02)00084-9 fatcat:x52nkophkfgtraxhuvtpqfvldu

Cross-view Transformers for real-time Map-view Semantic Segmentation [article]

Brady Zhou, Philipp Krähenbühl
2022 arXiv   pre-print
Each camera uses positional embeddings that depend on its intrinsic and extrinsic calibration.  ...  Our model is simple, easily parallelizable, and runs in real-time. The presented architecture performs at state-of-the-art on the nuScenes dataset, with 4x faster inference speeds.  ...  IIS-1845485 and IIS-2006820.  ... 
arXiv:2205.02833v1 fatcat:cqyrutnr7fcxrprkf2y6lefzki

A Simple Fix for Convolutional Neural Network via Coordinate Embedding [article]

Liliang Ren, Zhuonan Hao
2020 arXiv   pre-print
Our approach does not change the downstream model architecture and can be easily applied to the pre-trained models for the task like object detection.  ...  In this project, we proposed a simple approach to incorporate the coordinate information to the CNN model through coordinate embedding.  ...  As shown in Figure 3 , we construct the Coordinate Embedding (CoordEmb) using two trainable matrices X ∈ R H×W ×1 , Y ∈ R H×W ×1 to embed the x coordinate and y coordinate information of the input image  ... 
arXiv:2003.10589v1 fatcat:5eryejtembfxzckzngxrqd6qzu

Image Generators with Conditionally-Independent Pixel Synthesis [article]

Ivan Anokhin, Kirill Demochkin, Taras Khakhulin, Gleb Sterkin, Victor Lempitsky, Denis Korzhenkov
2020 arXiv   pre-print
Here, we present a new architecture for image generators, where the color value at each pixel is computed independently given the value of a random latent vector and the coordinate of that pixel.  ...  We analyze the modeling capabilities of such generators when trained in an adversarial fashion, and observe the new generators to achieve similar generation quality to state-of-the-art convolutional generators  ...  To mine the most similar faces, we extract faces using the MTCNN model [33] , and then compute their embeddings using FaceNet [24] (the public implementation of these models 4 was used).  ... 
arXiv:2011.13775v1 fatcat:kc3dgnzvnfcydfazsptqv4rnjm

Preface

Carlos Canal, Pascal Poizat, Mirko Viroli
2008 Electronical Notes in Theoretical Computer Science  
, António Ravara, Gwen Salaün, and Antonio Val-  ...  Program Committee We would like to thank also the referees that, in addition to the Program Committee, helped us in the review process: Maurizio Cimadamore, Francisco Durán, Paolo Milazzo, Razvan Popescu  ...  The article "Coordination Models Orc and Reo Compared" by José Proença and Dave Clarke takes a similar stance, and compares Reo and Orc by providing and analysing the two embeddings of one model into the  ... 
doi:10.1016/j.entcs.2008.03.095 fatcat:wx23wwqxizbabo3oujdk2gwoqe

PolyGen: An Autoregressive Generative Model of 3D Meshes [article]

Charlie Nash, Yaroslav Ganin, S. M. Ali Eslami, Peter W. Battaglia
2020 arXiv   pre-print
We present an approach which models the mesh directly, predicting mesh vertices and faces sequentially using a Transformer-based architecture.  ...  Existing learning-based approaches have avoided the challenges of working with 3D meshes, instead using alternative object representations that are more compatible with neural architectures and training  ...  Acknowledgements The authors thank Dan Rosenbaum, Sander Dieleman, Yujia Li and Craig Donner for useful discussions  ... 
arXiv:2002.10880v1 fatcat:jkkjcivfxnbcxhuyg4nhrmnhhq

Towards a Multi-modal, Multi-task Learning based Pre-training Framework for Document Representation Learning [article]

Subhojeet Pramanik, Shashank Mujumdar, Hima Patel
2022 arXiv   pre-print
We evaluate our framework on different standard document datasets and conduct exhaustive experiments to compare performance against various ablations of our framework and state-of-the-art baselines.  ...  Specifically, we introduce Document Topic Modelling and Document Shuffle Prediction as novel pre-training tasks to learn rich image representations along with the text and layout representations for documents  ...  compared to vanilla BERT architecture.  ... 
arXiv:2009.14457v2 fatcat:bx2aom56dbfhlb5qefe2vtx5im

Language Models with Transformers [article]

Chenguang Wang, Mu Li, Alexander J. Smola
2019 arXiv   pre-print
We propose Coordinate Architecture Search (CAS) to find an effective architecture through iterative refinement of the model.  ...  Recently, GPT and BERT demonstrate the efficacy of Transformer models on various NLP tasks using pre-trained language models on large-scale corpora.  ...  For BERT related model architectures, we use WordPiece embedding (Wu et al., 2016) to tokenize the training/validation/test split of the PTB, WT-2 and WT-103 respectively.  ... 
arXiv:1904.09408v2 fatcat:qpzmdn7v55b7zmyvhvoh76ejhu

Notice of Violation of IEEE Publication Principles A heterogeneous memory organization with minimum energy consumption in 3D chip-multiprocessors

Arghavan Asad, Salman Onsori, Mahmood Fathy, Mohammad Reza Jahed-Motlagh, Kaamran Raahemifar
2016 2016 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)  
Experimental results show that the proposed method improves energy-delay product (EDP) and performance by about 44.8% and 13.8% on average respectively compared with the traditional memory design where  ...  single technology is used.  ...  INTRODUCTION Chip multiprocessor (CMP) architectures have been widely used to meet growing demands on performance in embedded systems.  ... 
doi:10.1109/ccece.2016.7726817 dblp:conf/ccece/AsadOFMR16 fatcat:ej36y4vsi5e6zimwdh5dmirohe

Irregularly Tabulated MLP for Fast Point Feature Embedding [article]

Yusuke Sekikawa, Teppei Suzuki
2020 arXiv   pre-print
Aiming at drastic speedup for point-feature embeddings at test time, we propose a new framework that uses a pair of multi-layer perceptrons (MLP) and a lookup table (LUT) to transform point-coordinate  ...  LUTI-MLP also provides significant speedup for Jacobian computation of the embedding function wrt global pose coordinate on Lie algebra 𝔰𝔢(3) at test time, which could be used for point-set registration  ...  Fig. 1 . 1 LUTI-MLP embedding compared with MLP embedding for object classification. Top: MLP embedding used in PointNet [26]. Middle/Bottom: LUTI-MLP embedding at testing and training.  ... 
arXiv:2011.09852v1 fatcat:ymz7goe3tndibeyyyogowmqwoq

Notice of Violation of IEEE Publication Principles An Energy-Efficient Heterogeneous Memory Architecture for Future Dark Silicon Embedded Chip-Multiprocessors

Salman Onsori, Arghavan Asad, Kaamran Raahemifar, Mahmood Fathy
2018 IEEE Transactions on Emerging Topics in Computing  
In this article, we present a convex optimization model to design a 3D stacked hybrid memory architecture in order to minimize the future embedded systems energy consumption in the dark silicon era.  ...  Using conventional memory technologies in future designs in nanoscale era causes a drastic increase in leakage power consumption and temperature-related problems.  ...  Our approach uses 0-1 variables to specify the coordinates of each memory bank and processor core.  ... 
doi:10.1109/tetc.2016.2563323 fatcat:acioedfuo5fyvmkpulhaybadye

FRDet: Balanced and Lightweight Object Detector based on Fire-Residual Modules for Embedded Processor of Autonomous Driving [article]

Seontaek Oh, Ji-Hwan You, Young-Keun Kim
2020 arXiv   pre-print
For deployment on an embedded processor for autonomous driving, the object detection network should satisfy all of the accuracy, real-time inference, and light model size requirements.  ...  Therefore, we propose FRDet, a lightweight one-stage object detector that is balanced to satisfy all the constraints of accuracy, model size, and real-time processing on an embedded GPU processor for autonomous  ...  Also, SSDLite [27] , which is a widely used lightweight model for embedded GPUs, also is compared.  ... 
arXiv:2011.08061v1 fatcat:zkrhq7vivrccfiihgyfg474gta

Joint Image and 3D Shape Part Representation in Large Collections for Object Blending

Adrian Penate-Sanchez, Lourdes Agapito
2020 IEEE Access  
we are just unlucky and both estimations are incorrect.  ...  In many cases the 3D structure is unclear due to the perspective of an image, sometimes one of the parts cannot correctly estimate correctly the shape and that generates an incorrect blending and sometimes  ...  The experiment compares against the original results of Li [2] , Ghirdar [11] , HoG, AlexNet and Siamese networks. Test samples have not been used during the training of the EmbedN et architecture.  ... 
doi:10.1109/access.2020.2975106 fatcat:uxgdogrkdjgpvcy3z5kk3s23im

Unsupervised Feature Learning of Human Actions as Trajectories in Pose Embedding Manifold [article]

Jogendra Nath Kundu, Maharshi Gor, Phani Krishna Uppala, R. Venkatesh Babu
2018 arXiv   pre-print
Further, we use the pose embeddings generated by EnGAN to model human actions using a bidirectional RNN auto-encoder architecture, PoseRNN.  ...  An unsupervised human action modeling framework can provide useful pose-sequence representation, which can be utilized in a variety of pose analysis applications.  ...  Acknowledgements This work was supported by a CSIR Fellowship (Jogendra), and a project grant from Robert Bosch Centre for Cyber-Physical Systems, IISc.  ... 
arXiv:1812.02592v1 fatcat:37jnz4444faudnvaiao5d6sjym
« Previous Showing results 1 — 15 out of 157,074 results