140 Hits in 4.2 sec

Complaint-driven Training Data Debugging for Query 2.0 [article]

Weiyuan Wu, Lampros Flokas, Eugene Wu, Jiannan Wang
2020 pre-print
In response, we propose Rain, a complaint-driven training data debugging system.  ...  Debugging Query 2.0 is very challenging since an unexpected query result may be caused by the bugs in training data (e.g., wrong labels, corrupted features).  ...  The above use cases highlight desired criteria that motivate complaint-driven data debugging for Query 2.0. First, is the ability to express data errors at different points in the query pipeline.  ... 
doi:10.1145/3318464.3389696 arXiv:2004.05722v1 fatcat:md7phoi33fcd3iwkhaxi3ebsku

A survey and comparison of transformation tools based on the transformation tool contest

Edgar Jakumeit, Sebastian Buchwald, Dennis Wagelaar, Li Dan, Ábel Hegedüs, Markus Herrmannsdörfer, Tassilo Horn, Elina Kalnina, Christian Krause, Kevin Lano, Markus Lepper, Arend Rensink (+3 others)
2014 Science of Computer Programming  
Furthermore, we want to thank the reviewers for their valuable comments.  ...  Acknowledgements We want to thank Stephan Hildebrandt for his contributions to this article, the organizers of the Transformation Tool Contest for rendering this comparison possible, and the SHARE maintainers  ...  Being built for SQL like querying of graph repositories or databases, query languages comprise a direct match regarding the question for the means available for general data querying.  ... 
doi:10.1016/j.scico.2013.10.009 fatcat:rft2mq6z6zhulc6eetgbxuuuii

Interpreting Cloud Computer Vision Pain-Points: A Mining Study of Stack Overflow [article]

Alex Cummaudo, Rajesh Vasa, Scott Barnett, John Grundy, Mohamed Abdelrazek
2020 arXiv   pre-print
Such APIs look and feel like conventional APIs but abstract away data-driven probabilistic behaviour - the implications of a developer treating these APIs in the same way as other, traditional cloud services  ...  ACKNOWLEDGEMENTS We acknowledge Zac Brannelly, Fayaz Beigh, and Vivian Dabre for their assistance in categorising posts.  ...  [82] indicates the documentation does not adequately describe the training data for the API-developers do not know the required usage assumptions.  ... 
arXiv:2001.10130v1 fatcat:2cxh3hnaebb6fdan2bbjpbj7n4

A Roadmap for Big Model [article]

Sha Yuan, Hanyu Zhao, Shuai Zhao, Jiahong Leng, Yangxiao Liang, Xiaozhi Wang, Jifan Yu, Xin Lv, Zhou Shao, Jiaao He, Yankai Lin, Xu Han (+88 others)
2022 arXiv   pre-print
With the rapid development of deep learning, training Big Models (BMs) for multiple downstream tasks becomes a popular paradigm.  ...  We introduce 16 specific BM-related topics in those four parts, they are Data, Knowledge, Computing System, Parallel Training System, Language Model, Vision Model, Multi-modal Model, Theory&Interpretability  ...  However, there are constant complaints about its descriptive programming style and difficulties in debugging. Baidu proposed PaddlePaddle which is used for multi-level distributed training [233] .  ... 
arXiv:2203.14101v4 fatcat:rdikzudoezak5b36cf6hhne5u4

Design and evaluation of acceleration strategies for speeding up the development of dialog applications

Luis Fernando D'Haro, Ricardo de Córdoba, Rubén San-Segundo, Javier Ferreiros, José Manuel Pardo
2011 Speech Communication  
solving specific modality details such as confirmations of user answers or how to present them the lists of retrieved results after querying the backend database.  ...  several other accelerations such as configurable templates that can be used to define the different tasks in the service or the dialogs to obtain or show information to the user, automatic proposals for  ...  the Universidad Politécnica de Madrid, and to all the members of the GEMINI project for making possible the creation of the platform described in this paper.  ... 
doi:10.1016/j.specom.2011.05.008 fatcat:y7ugc6mtvzbl3cgcydkshd4wte

Origins of the D programming language

Walter Bright, Andrei Alexandrescu, Michael Parker
2020 Proceedings of the ACM on Programming Languages (PACMPL)  
Walter Bright, D's creator, is a mechanical engineer by education who started out working for Boeing designing gearboxes for the 757.  ...  This interest materialized into a C compiler, followed by compilers for C++, Java, and JavaScript.  ...  The charter of static if is code generation driven by introspection. Typically, generic code queries the parameterized types received and makes decisions depending on their capabilities.  ... 
doi:10.1145/3386323 fatcat:vlsjnroggbg2bk6hhhktrqafj4

Web data extraction, applications and techniques: A survey

Emilio Ferrara, Pasquale De Meo, Giacomo Fiumara, Robert Baumgartner
2014 Knowledge-Based Systems  
At the Enterprise level, Web Data Extraction techniques emerge as a key tool to perform data analysis in Business and Competitive Intelligence systems as well as for business process re-engineering.  ...  At the Social Web level, Web Data Extraction techniques allow to gather a large amount of structured data continuously generated and disseminated by Web 2.0, Social Media and Online Social Network users  ...  Among these systems we cite Virtuoso Sponger, 22 which generates Linked Data starting from different data sources and supports a wide range of data representation formats and Semantic Fire, 23 which  ... 
doi:10.1016/j.knosys.2014.07.007 fatcat:cb6zazpx7nfgxkmkiuoxqx5zyq

Understanding the past—illuminating the future

Nick Rushby, Jan Seabrook
2008 British Journal of Educational Technology  
Acknowledgements This project was funded by the British Educational Communications and Technology Agency and our thanks go to the staff of the Evidence and Evaluation Directorate for their support, initially  ...  In The projects covered a wide range of training requirements and applications, including:  an expert system on customer complaints procedure for use by catering students and small businesses  computer  ...  And Web 2.0 applications don't feature in research done in 2005...  ... 
doi:10.1111/j.1467-8535.2008.00816.x fatcat:wzqpvudo2zhwrggsxqswoq3g5u

CAPABLE D2.1: Requirements Table and Use Case Description

Mor Peleg, Giordano Lanzola, Silvia Panzarasa, Enea Parimbelli, Francesca Polce, Silvana Quaglini, Lucia Sacchi, Nicole Veggiotti, Alexandra Kogan, Roy Leizer, Matteo Gabetta, Ronald Cornet (+20 others)
2020 Zenodo  
Following our iterative development approach for the CAPABLE system, this document presents the majority of the functional and non-functional requirements.  ...  requirements include clinical requirements regarding the clinical scope of decision-support and its content (knowledge base), patient and clinician user needs and requirements, and technical system requirements for  ...  KDOM will query the Data Platform for the needed data to calculate the abstraction.  ... 
doi:10.5281/zenodo.4540451 fatcat:vusuh5dh2ze7bk4nlsgdysi3fy

D6.4: Development of Prototypal New Services

Janez Povh
2019 Zenodo  
Additionally, we have analysed within PRACE-5IP project two new services: (5) Evaluation of lightweight virtualisation technology and (6) Evaluation of new prototypes for Data Analytics services.  ...  in many test scripts, benchmark results, dissemination and training activities and are described in this document.  ...  was used for debugging Lustre at large scale.  ... 
doi:10.5281/zenodo.6805985 fatcat:lhb6v3nen5daxpydswfonlkqxu

Verilog HDL and its ancestors and descendants

Peter Flake, Phil Moorby, Steve Golson, Arturo Salz, Simon Davidmann
2020 Proceedings of the ACM on Programming Languages (PACMPL)  
For large-scale digital logic design, previous schematic-based techniques have transformed into textual registertransfer level (RTL) descriptions written in Verilog.  ...  A related complaint is that BSV's capability for interfacing to other HDLs such as VHDL or Verilog is extremely limited.  ...  Debugging a Design Language features for verifying that a design is correct, and if not, for debugging it to find out what is wrong, were built into the same Verilog syntax as the rest of the language.  ... 
doi:10.1145/3386337 fatcat:ttezkcr6pzgppbeofpi23vu2wy

"A veritable bucket of facts" origins of the data base management system

Thomas Haigh
2006 SIGMOD record  
Bertossi is Faculty Fellow of IBM Center for Advanced Studies (Toronto Lab.). Several contributions and comments by Loreto Bravo are very much appreciated.  ...  Acknowledgements The author would like to thank Jef Wijsen and Alain Pirotte for their insightful comments that allowed to improve the quality of the paper. Acknowledgement.  ...  In particular, the complexity of relational query construction meant that to query and update the data base still required the involvement of a programmer, a specially written application program, or trained  ... 
doi:10.1145/1147376.1147382 fatcat:pk5odqlhqnctdepmcrfsriputi

The Tools Perspective on Software Reverse Engineering [chapter]

Holger M. Kienle, Hausi A. Müller
2010 Advances in Computers  
This chapter explores the issue of building tools for reverse engineering.  ...  The purpose of each lens is to focus on a critical topic for tool building by surveying the current state-of-the-art and identifying challenges that need to be addressed.  ...  In order to accommodate third-party tools, SolidFX offers control-integration with a query API to the fact database and data-integration based on various formats (XML for ASTs, XMI for UML diagrams, SQL  ... 
doi:10.1016/s0065-2458(10)79005-7 fatcat:xc7i6ccaqjdjrgwf2itakyjcwa

On the Opportunities and Risks of Foundation Models [article]

Rishi Bommasani, Drew A. Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S. Bernstein, Jeannette Bohg, Antoine Bosselut, Emma Brunskill, Erik Brynjolfsson, Shyamal Buch (+102 others)
2022 arXiv   pre-print
., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks.  ...  procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic and environmental impact, legal  ...  ., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic  ... 
arXiv:2108.07258v3 fatcat:kohwrwk2ybf7fd7wsuz2gp65ki

The Fourth Paradigm – Data-Intensive Scientific Discovery [chapter]

Tony Hey
2012 Communications in Computer and Information Science  
The F o u r T H Data-In t e n s I v e scIen t I f I c DIs c o v e r y P a r a d i g m MiCroSoF T reSe arCH REDMOND, WASHINGTON  ...  And, of course, we thank Jim Gray, for inspiring us. ACKN OWLE DG M E NTS  ...  We also thank our families and colleagues for their support during the intensive editorial process.  ... 
doi:10.1007/978-3-642-33299-9_1 fatcat:etue636ubrandga5corvkx3d7u
« Previous Showing results 1 — 15 out of 140 results