Amund Tveit | Microsoft - Academia.edu (original) (raw)
Papers by Amund Tveit
In this paper we present DeepLearningKit-an open source framework that supports using pre-trained... more In this paper we present DeepLearningKit-an open source framework that supports using pre-trained deep learning models (convolutional neural networks) for iOS, OS X and tvOS. DeepLearningKit is developed in Metal in order to utilize the GPU efficiently and Swift for integration with applications, e.g. iOS-based mobile apps on iPhone/iPad, tvOS-based apps for the big screen, or OS X desktop applications. The goal is to support using deep learning models trained with popular frameworks such as Caffe, Torch, TensorFlow, Theano, Pylearn, Deeplearning4J and Mocha. Given the massive GPU resources and time required to train Deep Learning models we suggest an App Store like model to distribute and download pretrained and reusable Deep Learning models.
Proceedings of the International …
This paper gives an overview of recent cluster computing technology trends with a focus on cluste... more This paper gives an overview of recent cluster computing technology trends with a focus on clusters that are used for high-performance computing (HPC). Here, a network of workstations and/or PCs is used rather than a traditional supercomputer. Although the network speed is still generally slow compared to available processing speed, and implementing distributed memory programs is generally harder than shared-memory programs, the price and flexibility of such systems cannot be underestimated. Tools like ...
ABSTRACT This chapter considers an application of software agents to mobile commerce services pro... more ABSTRACT This chapter considers an application of software agents to mobile commerce services provision. With the increasing number of e-commerce services for mobile devices, there are challenges in making these services more personalized and in taking into account the severely constrained bandwidth and restricted user interface these devices currently provide. In this chapter we present an agent-based platform for support of mobile commerce usingwireless devices.
Abstract This study has been performed in order to categorize and measure usage of different info... more Abstract This study has been performed in order to categorize and measure usage of different information sources and types in a well defined stage of clinical work. The underlying motivation is to improve computer-supported presentation and retrieval of relevant information and to be able to evaluate the functionality of a future improved interface to the electronic patient record (EPR).
With the increasing number of competing mobile services, the process of acquiring new, and retain... more With the increasing number of competing mobile services, the process of acquiring new, and retaining existing customers becomes harder. Acquisition of new customers can, to some degree, be done with increased marketing efforts, but marketing has less effect for retaining existing customers, since they allready know the service. Improving the mobile service is likely to be the best way to retain existing customers.
Abstract Software agent research is the most recent and prominent contribution to the field of so... more Abstract Software agent research is the most recent and prominent contribution to the field of software engineering. Nevertheless, the process of finding relevant litterature in this domain is a tremendous challenge, and it is frequently not found using traditional Internet search engines. This paper meets this challenge by applying other and usually better information sources, examples of such sources are library and topic-specific databases.
Abstract This paper suggests an experimental approach of how to apply inductive logic programming... more Abstract This paper suggests an experimental approach of how to apply inductive logic programming in the discovery of web usage patterns in the form of first-order rules representing user sessions. Such rules may be used to improve the quality and the performance of a web site. The experiment has been done using the Progol Inductive Logic Programming System, and the data source are log files from an Apache web server. Keywords: Inductive Logic Programming, Data Mining
How to cope with information overload is becoming an increasingly important problem even for scie... more How to cope with information overload is becoming an increasingly important problem even for scientists. Search engines such as Scholar, CiteSeer, SmealSearch, Google, MSN and Yahoo tries to solve this problem by indexing (variable size) samples of publicly available texts (in various formats) from the web. With the recent introduction of desktop search the user's own data so that search can be performed simultaneously locally on and on the web.
Abstract Cyberspace plays an increasingly important role in people's life due to its plentiful of... more Abstract Cyberspace plays an increasingly important role in people's life due to its plentiful offering of services and information, eg the Word Wide Web, the Mobile Web and Online Games. However, the usability of cyberspace services is frequently reduced by its lack of customization according to individual needs and preferences.
Proceedings of the …, Jan 1, 2003
This paper gives an overview of recent cluster computing technology trends with a focus on cluste... more This paper gives an overview of recent cluster computing technology trends with a focus on clusters that are used for high-performance computing (HPC). Here, a network of workstations and/or PCs is used rather than a traditional supercomputer. Although the network speed is still generally slow compared to available processing speed, and implementing distributed memory programs is generally harder than shared-memory programs, the price and flexibility of such systems cannot be underestimated. Tools like ...
Knowledge-Based …, Jan 1, 2005
Abstract: The most used method of finding logical rules from data, inductive logic programming (I... more Abstract: The most used method of finding logical rules from data, inductive logic programming (ILP), has shown successful, but unfortunately not very scalable with increasing problem size. In this report a model for doing induction of logical rules, using the concepts of the potentially more scalable method of genetic algorithm, is suggested.
In this paper we present DeepLearningKit-an open source framework that supports using pre-trained... more In this paper we present DeepLearningKit-an open source framework that supports using pre-trained deep learning models (convolutional neural networks) for iOS, OS X and tvOS. DeepLearningKit is developed in Metal in order to utilize the GPU efficiently and Swift for integration with applications, e.g. iOS-based mobile apps on iPhone/iPad, tvOS-based apps for the big screen, or OS X desktop applications. The goal is to support using deep learning models trained with popular frameworks such as Caffe, Torch, TensorFlow, Theano, Pylearn, Deeplearning4J and Mocha. Given the massive GPU resources and time required to train Deep Learning models we suggest an App Store like model to distribute and download pretrained and reusable Deep Learning models.
Proceedings of the International …
This paper gives an overview of recent cluster computing technology trends with a focus on cluste... more This paper gives an overview of recent cluster computing technology trends with a focus on clusters that are used for high-performance computing (HPC). Here, a network of workstations and/or PCs is used rather than a traditional supercomputer. Although the network speed is still generally slow compared to available processing speed, and implementing distributed memory programs is generally harder than shared-memory programs, the price and flexibility of such systems cannot be underestimated. Tools like ...
ABSTRACT This chapter considers an application of software agents to mobile commerce services pro... more ABSTRACT This chapter considers an application of software agents to mobile commerce services provision. With the increasing number of e-commerce services for mobile devices, there are challenges in making these services more personalized and in taking into account the severely constrained bandwidth and restricted user interface these devices currently provide. In this chapter we present an agent-based platform for support of mobile commerce usingwireless devices.
Abstract This study has been performed in order to categorize and measure usage of different info... more Abstract This study has been performed in order to categorize and measure usage of different information sources and types in a well defined stage of clinical work. The underlying motivation is to improve computer-supported presentation and retrieval of relevant information and to be able to evaluate the functionality of a future improved interface to the electronic patient record (EPR).
With the increasing number of competing mobile services, the process of acquiring new, and retain... more With the increasing number of competing mobile services, the process of acquiring new, and retaining existing customers becomes harder. Acquisition of new customers can, to some degree, be done with increased marketing efforts, but marketing has less effect for retaining existing customers, since they allready know the service. Improving the mobile service is likely to be the best way to retain existing customers.
Abstract Software agent research is the most recent and prominent contribution to the field of so... more Abstract Software agent research is the most recent and prominent contribution to the field of software engineering. Nevertheless, the process of finding relevant litterature in this domain is a tremendous challenge, and it is frequently not found using traditional Internet search engines. This paper meets this challenge by applying other and usually better information sources, examples of such sources are library and topic-specific databases.
Abstract This paper suggests an experimental approach of how to apply inductive logic programming... more Abstract This paper suggests an experimental approach of how to apply inductive logic programming in the discovery of web usage patterns in the form of first-order rules representing user sessions. Such rules may be used to improve the quality and the performance of a web site. The experiment has been done using the Progol Inductive Logic Programming System, and the data source are log files from an Apache web server. Keywords: Inductive Logic Programming, Data Mining
How to cope with information overload is becoming an increasingly important problem even for scie... more How to cope with information overload is becoming an increasingly important problem even for scientists. Search engines such as Scholar, CiteSeer, SmealSearch, Google, MSN and Yahoo tries to solve this problem by indexing (variable size) samples of publicly available texts (in various formats) from the web. With the recent introduction of desktop search the user's own data so that search can be performed simultaneously locally on and on the web.
Abstract Cyberspace plays an increasingly important role in people's life due to its plentiful of... more Abstract Cyberspace plays an increasingly important role in people's life due to its plentiful offering of services and information, eg the Word Wide Web, the Mobile Web and Online Games. However, the usability of cyberspace services is frequently reduced by its lack of customization according to individual needs and preferences.
Proceedings of the …, Jan 1, 2003
This paper gives an overview of recent cluster computing technology trends with a focus on cluste... more This paper gives an overview of recent cluster computing technology trends with a focus on clusters that are used for high-performance computing (HPC). Here, a network of workstations and/or PCs is used rather than a traditional supercomputer. Although the network speed is still generally slow compared to available processing speed, and implementing distributed memory programs is generally harder than shared-memory programs, the price and flexibility of such systems cannot be underestimated. Tools like ...
Knowledge-Based …, Jan 1, 2005
Abstract: The most used method of finding logical rules from data, inductive logic programming (I... more Abstract: The most used method of finding logical rules from data, inductive logic programming (ILP), has shown successful, but unfortunately not very scalable with increasing problem size. In this report a model for doing induction of logical rules, using the concepts of the potentially more scalable method of genetic algorithm, is suggested.
This presentation will give an overview of mapreduce-based algorithms described in recent papers ... more This presentation will give an overview of mapreduce-based algorithms described in recent papers written by academic and industrial researchers. Included areas: AI/Machine Learning, Bioinformatics, Information Retrieval. Focus will be on patterns of problems and the corresponding mapreduce solution patterns. Some background material: http://mapreducepatterns.org