data representation
Recently Published Documents





2022 ◽  
Vol 13 (1) ◽  
pp. 1-25
Yuandong Wang ◽  
Hongzhi Yin ◽  
Tong Chen ◽  
Chunyang Liu ◽  
Ben Wang ◽  

In recent years, ride-hailing services have been increasingly prevalent, as they provide huge convenience for passengers. As a fundamental problem, the timely prediction of passenger demands in different regions is vital for effective traffic flow control and route planning. As both spatial and temporal patterns are indispensable passenger demand prediction, relevant research has evolved from pure time series to graph-structured data for modeling historical passenger demand data, where a snapshot graph is constructed for each time slot by connecting region nodes via different relational edges (origin-destination relationship, geographical distance, etc.). Consequently, the spatiotemporal passenger demand records naturally carry dynamic patterns in the constructed graphs, where the edges also encode important information about the directions and volume (i.e., weights) of passenger demands between two connected regions. aspects in the graph-structure data. representation for DDW is the key to solve the prediction problem. However, existing graph-based solutions fail to simultaneously consider those three crucial aspects of dynamic, directed, and weighted graphs, leading to limited expressiveness when learning graph representations for passenger demand prediction. Therefore, we propose a novel spatiotemporal graph attention network, namely Gallat ( G raph prediction with all at tention) as a solution. In Gallat, by comprehensively incorporating those three intrinsic properties of dynamic directed and weighted graphs, we build three attention layers to fully capture the spatiotemporal dependencies among different regions across all historical time slots. Moreover, the model employs a subtask to conduct pretraining so that it can obtain accurate results more quickly. We evaluate the proposed model on real-world datasets, and our experimental results demonstrate that Gallat outperforms the state-of-the-art approaches.

2022 ◽  
Vol 15 (1) ◽  
pp. 1-21
Chen Wu ◽  
Mingyu Wang ◽  
Xinyuan Chu ◽  
Kun Wang ◽  
Lei He

Low-precision data representation is important to reduce storage size and memory access for convolutional neural networks (CNNs). Yet, existing methods have two major limitations: (1) requiring re-training to maintain accuracy for deep CNNs and (2) needing 16-bit floating-point or 8-bit fixed-point for a good accuracy. In this article, we propose a low-precision (8-bit) floating-point (LPFP) quantization method for FPGA-based acceleration to overcome the above limitations. Without any re-training, LPFP finds an optimal 8-bit data representation with negligible top-1/top-5 accuracy loss (within 0.5%/0.3% in our experiments, respectively, and significantly better than existing methods for deep CNNs). Furthermore, we implement one 8-bit LPFP multiplication by one 4-bit multiply-adder and one 3-bit adder, and therefore implement four 8-bit LPFP multiplications using one DSP48E1 of Xilinx Kintex-7 family or DSP48E2 of Xilinx Ultrascale/Ultrascale+ family, whereas one DSP can implement only two 8-bit fixed-point multiplications. Experiments on six typical CNNs for inference show that on average, we improve throughput by over existing FPGA accelerators. Particularly for VGG16 and YOLO, compared to six recent FPGA accelerators, we improve average throughput by 3.5 and 27.5 and average throughput per DSP by 4.1 and 5 , respectively.

Nagireddy Kavya

Abstract: In this paper, we present the design and implementation of Floating point addition and Floating point Multiplication. There are many multipliers in existence in which Floating point Multiplication and Floating point addition offers a high precision and more accuracy for the data representation of the image. This project is designed and simulated on Xilinx ISE 14.7 version software using verilog. Simulation results show area reduction and delay reduction as compared to the conventional method. Keywords: FIR Filter, Floating point Addition, Floating point Multiplication, Carry Look Ahead Adder

2022 ◽  
pp. 1-38
William Paul ◽  
Armin Hadzic ◽  
Neil Joshi ◽  
Fady Alajaji ◽  
Philippe Burlina

Abstract We propose a novel method for enforcing AI fairness with respect to protected or sensitive factors. This method uses a dual strategy performing training and representation alteration (TARA) for the mitigation of prominent causes of AI bias. It includes the use of representation learning alteration via adversarial independence to suppress the bias-inducing dependence of the data representation from protected factors and training set alteration via intelligent augmentation to address bias-causing data imbalance by using generative models that allow the fine control of sensitive factors related to underrepresented populations via domain adaptation and latent space manipulation. When testing our methods on image analytics, experiments demonstrate that TARA significantly or fully debiases baseline models while outperforming competing debiasing methods that have the same amount of information—for example, with (% overall accuracy, % accuracy gap) = (78.8, 0.5) versus the baseline method's score of (71.8, 10.5) for Eye-PACS, and (73.7, 11.8) versus (69.1, 21.7) for CelebA. Furthermore, recognizing certain limitations in current metrics used for assessing debiasing performance, we propose novel conjunctive debiasing metrics. Our experiments also demonstrate the ability of these novel metrics in assessing the Pareto efficiency of the proposed methods.

Semantic Web ◽  
2022 ◽  
pp. 1-34
Fajar J. Ekaputra ◽  
Andreas Ekelhart ◽  
Rudolf Mayer ◽  
Tomasz Miksa ◽  
Tanja Šarčević ◽  

Small and medium-sized organisations face challenges in acquiring, storing and analysing personal data, particularly sensitive data (e.g., data of medical nature), due to data protection regulations, such as the GDPR in the EU, which stipulates high standards in data protection. Consequently, these organisations often refrain from collecting data centrally, which means losing the potential of data analytics and learning from aggregated user data. To enable organisations to leverage the full-potential of the collected personal data, two main technical challenges need to be addressed: (i) organisations must preserve the privacy of individual users and honour their consent, while (ii) being able to provide data and algorithmic governance, e.g., in the form of audit trails, to increase trust in the result and support reproducibility of the data analysis tasks performed on the collected data. Such an auditable, privacy-preserving data analysis is currently challenging to achieve, as existing methods and tools only offer partial solutions to this problem, e.g., data representation of audit trails and user consent, automatic checking of usage policies or data anonymisation. To the best of our knowledge, there exists no approach providing an integrated architecture for auditable, privacy-preserving data analysis. To address these gaps, as the main contribution of this paper, we propose the WellFort approach, a semantic-enabled architecture for auditable, privacy-preserving data analysis which provides secure storage for users’ sensitive data with explicit consent, and delivers a trusted, auditable analysis environment for executing data analytic processes in a privacy-preserving manner. Additional contributions include the adaptation of Semantic Web technologies as an integral part of the WellFort architecture, and the demonstration of the approach through a feasibility study with a prototype supporting use cases from the medical domain. Our evaluation shows that WellFort enables privacy preserving analysis of data, and collects sufficient information in an automated way to support its auditability at the same time.

2022 ◽  
Vol 20 (8) ◽  
pp. 3119
O. V. Kopylova ◽  
A. I. Ershova ◽  
M. S. Pokrovskaya ◽  
A. N. Meshkov ◽  
I. A. Efimova ◽  

Aim. To analyze the structure of clinical data, as well as the principles of collecting and storing related data of the biobank of the National Medical Research Center for Therapy and Preventive Medicine (hereinafter Biobank).Material and methods. The analysis was carried out using the documentation available in the Biobank, as well as the databases used in its work. The paper presents clinical data on biosamples available in the Biobank as of August 18, 2021.Results. At the time of analysis, the Biobank had 373547 samples collected from 54192 patients within 37 research projects. The article presents the analysis of data representation and quantitative assessment of the presence/absence of common diagnoses in clinical projects. Approaches to documenting clinical information associated with biological samples stored in the Biobank were assessed. The methods and tools used for standardization and automation of processes used in the Biobank were substantiated.Conclusion. The Biobank of the National Medical Research Center for Therapy and Preventive Medicine is the largest research biobank in Russia, which meets all modern international requirements and is one of the key structures that improve the research quality and intensify their conduct both within the one center and in cooperation with other biobanks and scientific institutions. The collection and systematic storage of clinical abstracts of biological samples is an integral and most important part of the Biobank’s work.

10.2196/30557 ◽  
2022 ◽  
Vol 10 (1) ◽  
pp. e30557
Aditya Vaidyam ◽  
John Halamka ◽  
John Torous

Background There is a growing need for the integration of patient-generated health data (PGHD) into research and clinical care to enable personalized, preventive, and interactive care, but technical and organizational challenges, such as the lack of standards and easy-to-use tools, preclude the effective use of PGHD generated from consumer devices, such as smartphones and wearables. Objective This study outlines how we used mobile apps and semantic web standards such as HTTP 2.0, Representational State Transfer, JSON (JavaScript Object Notation), JSON Schema, Transport Layer Security (version 1.3), Advanced Encryption Standard-256, OpenAPI, HTML5, and Vega, in conjunction with patient and provider feedback to completely update a previous version of mindLAMP. Methods The Learn, Assess, Manage, and Prevent (LAMP) platform addresses the abovementioned challenges in enhancing clinical insight by supporting research, data analysis, and implementation efforts around PGHD as an open-source solution with freely accessible and shared code. Results With a simplified programming interface and novel data representation that captures additional metadata, the LAMP platform enables interoperability with existing Fast Healthcare Interoperability Resources–based health care systems as well as consumer wearables and services such as Apple HealthKit and Google Fit. The companion Cortex data analysis and machine learning toolkit offer robust support for artificial intelligence, behavioral feature extraction, interactive visualizations, and high-performance data processing through parallelization and vectorization techniques. Conclusions The LAMP platform incorporates feedback from patients and clinicians alongside a standards-based approach to address these needs and functions across a wide range of use cases through its customizable and flexible components. These range from simple survey-based research to international consortiums capturing multimodal data to simple delivery of mindfulness exercises through personalized, just-in-time adaptive interventions.

2022 ◽  
pp. 913-932
G. Vimala Kumari ◽  
G. Sasibhushana Rao ◽  
B. Prabhakara Rao

This article presents an image compression method using feed-forward back-propagation neural networks (NNs). Marked progress has been made in the area of image compression in the last decade. Image compression removing redundant information in image data is a solution for storage and data transmission problems for huge amounts of data. NNs offer the potential for providing a novel solution to the problem of image compression by its ability to generate an internal data representation. A comparison among various feed-forward back-propagation training algorithms was presented with different compression ratios and different block sizes. The learning methods, the Levenberg Marquardt (LM) algorithm and the Gradient Descent (GD) have been used to perform the training of the network architecture and finally, the performance is evaluated in terms of MSE and PSNR using medical images. The decompressed results obtained using these two algorithms are computed in terms of PSNR and MSE along with performance plots and regression plots from which it can be observed that the LM algorithm gives more accurate results than the GD algorithm.

Sign in / Sign up

Export Citation Format

Share Document