Platform for Open Development of Systems of Artificial Intelligence

Risultati finali

Conceived as a dissemination tool to interest stakeholders and the public in the project. Project partners will distribute it at various events.

Bonseyes video

Conceived as a dissemination tool to interest researchers, stakeholders and the public in the project, also addressing an audience which can not easily be reached via other means. The video will be made available on the Bonseyes website and channels such as youtube.

Website

Public project website serving as a dissemination gateway.

Revised AI Marketplace

Revised Marketplace development based on the results from Open Developer Community Validation.

Validation Report

Documentation of the Bonseyes use case validation results and of the open developer community validation results.

Data Marketplace Report

Documentation of the data marketplace design and implementation.

Demonstrator Proof of Concepts

Report detailing demonstrator achievements.

Initial Platform Deployment Methods and Tools
Initial Deep Learning Methods
Revised Deep Learning Methods
Revised Platform Deployment Methods and Tools

Pubblicazioni

Privacy and Trust in Cloud-Based Marketplaces for AI and Data Resources

Autori: Ahmadi , Vida; Tutschku , Kurt
Pubblicato in: IFIP Advances in Information and Communication Technology, Numero 5, 2017, Pagina/e 223-225, ISBN 3319-591703
Editore: Springer

Performance Analysis and Optimization of Sparse Matrix-Vector Multiplication on Modern Multi- and Many-Core Processors

Autori: Athena Elafrou, Georgios Goumas, Nectarios Koziris
Pubblicato in: 2017 46th International Conference on Parallel Processing (ICPP), 2017, Pagina/e 292-301, ISBN 978-1-5386-1042-8
Editore: IEEE
DOI: 10.1109/ICPP.2017.38

Optimal DNN primitive selection with partitioned boolean quadratic programming

Autori: Andrew Anderson, David Gregg
Pubblicato in: Proceedings of the 2018 International Symposium on Code Generation and Optimization - CGO 2018, 2018, Pagina/e 340-351, ISBN 9781-450356176
Editore: ACM Press
DOI: 10.1145/3179541.3168805

Pricing of Data Products in Data Marketplaces

Autori: Samuel A. Fricker, Yuliyan V. Maksimov
Pubblicato in: Lecture Notes in Business Information Processing (LNBIP), Numero 304, 2017, Pagina/e 49-66, ISBN 978-3-319-69190-9
Editore: Springer International Publishing
DOI: 10.1007/978-3-319-69191-6_4

Characterising Across-Stack Optimisations for Deep Convolutional Neural Networks

Autori: Jack Turner, Jose Cano, Valentin Radu, Elliot J. Crowley, Michael O'Boyle, Amos Storkey
Pubblicato in: 2018 IEEE International Symposium on Workload Characterization (IISWC), 2018, Pagina/e 101-110, ISBN 978-1-5386-6780-4
Editore: IEEE
DOI: 10.1109/IISWC.2018.8573503

QUENN - QUantization engine for low-power neural networks

Autori: Miguel de Prado, Maurizio Denna, Luca Benini, Nuria Pazos
Pubblicato in: Proceedings of the 15th ACM International Conference on Computing Frontiers - CF '18, 2018, Pagina/e 36-44, ISBN 9781-450357616
Editore: ACM Press
DOI: 10.1145/3203217.3203282

Towards Privacy Requirements for Collaborative Development of AI Applications

Autori: Ahmadi Mehri, Vida; Ilie, Dragos; Tutschku, Kurt
Pubblicato in: Numero 1, 2018
Editore: BTH

Learning to infer: RL-based search for DNN primitive selection on Heterogeneous Embedded Systems

Autori: de Prado, Miguel; Pazos, Nuria; Benini, Luca
Pubblicato in: Numero 1, 2018
Editore: ArXiv

Distilling with Performance Enhanced Students

Autori: Turner, Jack; Crowley, Elliot J.; Radu, Valentin; Cano, José; Storkey, Amos; O'Boyle, Michael
Pubblicato in: Numero 1, 2018
Editore: Proceedings of 27th International Conference on Artificial Neural Networks

On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length

Autori: Jastrzębski, Stanislaw; Kenton, Zachary; Ballas, Nicolas; Fischer, Asja; Bengio, Yoshua; Storkey, Amos
Pubblicato in: Jastrzębski , S , Kenton , Z , Ballas , N , Fischer , A , Bengio , Y & Storkey , A 2019 , ' On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length ' , Paper presented at Seventh International Conference on Learning Representations , New Orleans , United States , 6/05/19 - 9/05/19 ., Numero 1, 2018
Editore: Seventh International Conference on Learning Representations (2019)

DNN's Sharpest Directions Along the SGD Trajectory

Autori: Stanisław Jastrzębski, Zachary Kenton, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos Storkey
Pubblicato in: 2018
Editore: ArXiv

Three Factors Influencing Minima in SGD

Autori: Stanisław Jastrzębski, Zachary Kenton, Devansh Arpit, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos Storkey
Pubblicato in: 2018
Editore: ArXiv

Privacy and DRM Requirements for Collaborative Development of AI Applications

Autori: Vida Ahmadi Mehri, Dragos Ilie, Kurt Tutschku
Pubblicato in: Proceedings of the 13th International Conference on Availability, Reliability and Security - ARES 2018, 2018, Pagina/e 1-8, ISBN 9781-450364485
Editore: ACM Press
DOI: 10.1145/3230833.3233268

Designing a Secure IoT System Architecture from a Virtual Premise for a Collaborative AI Lab

Autori: Mehri, V. A., Ilie, D. & Tutschku, K.
Pubblicato in: 2019
Editore: Workshop on Decentralized IoT Systems and Security (DISS) 2019

Framework for Analysis of Multi-party Collaboration

Autori: Yuliyan V. Maksimov, Samuel A. Fricker
Pubblicato in: 2019 IEEE 27th International Requirements Engineering Conference Workshops (REW), 2019, Pagina/e 44-53, ISBN 978-1-7281-5165-6
Editore: IEEE
DOI: 10.1109/rew.2019.00013

Distributed Ledger for Provenance Tracking of Artificial Intelligence Assets

Autori: Philipp Lüthi, Thibault Gagnaux, Marcel Gygli
Pubblicato in: 2019
Editore: ArXiv

Scalar Arithmetic Multiple Data: Customizable Precision for Deep Neural Networks

Autori: Andrew Anderson, Michael Doyle, David Gregg
Pubblicato in: 2019 IEEE 26th Symposium on Computer Arithmetic (ARITH), 2019, Pagina/e 61-68, ISBN 978-1-7281-3366-9
Editore: IEEE
DOI: 10.1109/arith.2019.00018

Performance-Oriented Neural Architecture Search

Autori: Andrew Anderson, Jing Su, Rozenn Dahyot, David Gregg
Pubblicato in: 2020
Editore: ArXiv

BlockSwap: Fisher-guided Block Substitution for Network Compression on a Budget

Autori: Jack Turner, Elliot J. Crowley, Michael O'Boyle, Amos Storkey, Gavin Gray
Pubblicato in: 2020
Editore: ArXiv

Performance Aware Convolutional Neural Network Channel Pruning for Embedded GPUs

Autori: Valentin Radu, Kuba Kaszyk, Yuan Wen, Jack Turner, Jose Cano, Elliot J. Crowley, Bjorn Franke, Amos Storkey, Michael O'Boyle
Pubblicato in: 2019
Editore: ArXiv

IoT meets distributed AI - Deployment scenarios of Bonseyes AI applications on FIWARE

Autori: Lucien Moor, Lukas Bitter, Miguel De Prado, Nuria Pazos, Nabil Ouerhani
Pubblicato in: 2019 IEEE 38th International Performance Computing and Communications Conference (IPCCC), 2019, Pagina/e 1-2, ISBN 978-1-7281-1025-7
Editore: Piscataway, New Jersey, USA
DOI: 10.1109/ipccc47392.2019.8958742

How to train your MAML

Autori: Antoniou, Antreas; Edwards, Harrison; Storkey, Amos
Pubblicato in: Seventh International Conference on Learning Representations, 2019
Editore: ICLR 2019

BONSEYES - Platform for Open Development of Systems of Artificial Intelligence: Invited paper

Autori: Tim Llewellynn, Sebastian Koller, Georgios Goumas, Peter Leitner, Ganesh Dasika, Lei Wang, Kurt Tutschku, M. Milagro Fern?ndez-Carrobles, Oscar Deniz, Samuel Fricker, Amos Storkey, Nuria Pazos, Gordana Velikic, Kirsten Leufgen, Rozenn Dahyot
Pubblicato in: Proceedings of the Computing Frontiers Conference on ZZZ - CF'17, 2017, Pagina/e 299-304, ISBN 9781-450344876
Editore: ACM Press
DOI: 10.1145/3075564.3076259

Parallel Multi Channel convolution using General Matrix Multiplication

Autori: Aravind Vasudevan, Andrew Anderson, David Gregg
Pubblicato in: 2017 IEEE 28th International Conference on Application-specific Systems, Architectures and Processors (ASAP), 2017, Pagina/e 19-24, ISBN 978-1-5090-4825-0
Editore: IEEE
DOI: 10.1109/ASAP.2017.7995254

Flexible Privacy and High Trust in the Next Generation Internet - The Use Case of a Cloud-based Marketplace for AI

Autori: Mehri, Vida. A., Tutschku, Kurt
Pubblicato in: SNCNW - Swedish National Computer Networking Workshop, Halmstad, 2017
Editore: Halmstad university

Low-memory GEMM-based convolution algorithms for deep neural networks

Autori: Anderson, Andrew; Vasudevan, Aravind; Keane, Cormac; Gregg, David
Pubblicato in: Numero 2, 2017
Editore: ArXiv

Moonshine: Distilling with Cheap Convolutions

Autori: Crowley, Elliot J.; Gray, Gavin; Storkey, Amos
Pubblicato in: Numero 1, 2017
Editore: arXiv

Accelerating Deep Neural Networks on Low Power Heterogeneous Architectures.

Autori: Loukadakis, M., Cano, J. & O’Boyle, M.
Pubblicato in: 11th International Workshop on Programmability and Architectures for Heterogeneous Multicores (MULTIPROG-2018). 11th International Workshop on Programmability and Architectures for Heterogeneous Multicores (MULTIPROG-2018), Manchester, United Kingdom, 24 January, 2018
Editore: -

Separable Layers Enable Structured Efficient Linear Substitutions

Autori: Gray, Gavin; Crowley, Elliot J.; Storkey, Amos
Pubblicato in: Numero 1, 2019
Editore: ArXiv

AI Pipeline - bringing AI to you. End-to-end integration of data, algorithms and deployment tools

Autori: de Prado, Miguel; Su, Jing; Dahyot, Rozenn; Saeed, Rabia; Keller, Lorenzo; Vallez, Noelia
Pubblicato in: Numero 1, 2019
Editore: ArXiv

Assume, Augment and Learn: Unsupervised Few-Shot Meta-Learning via Random Labels and Data Augmentation

Autori: Antoniou, Antreas; Storkey, Amos
Pubblicato in: Numero 1, 2019
Editore: ArXiv

A Closer Look at Structured Pruning for Neural Network Compression

Autori: Crowley, Elliot J.; Turner, Jack; Storkey, Amos; O'Boyle, Michael
Pubblicato in: Numero 1, 2019
Editore: ArXiv

RecNets: Channel-wise Recurrent Convolutional Neural Networks

Autori: Retsinas, G., Elafrou, A., Goumas, G. & Maragos, P.
Pubblicato in: 2020
Editore: arXiv.org

Artifact Compatibility for Enabling Collaboration in the Artificial Intelligence Ecosystem

Autori: Yuliyan V. Maksimov, Samuel A. Fricker, Kurt Tutschku
Pubblicato in: Software Business - 9th International Conference, ICSOB 2018, Tallinn, Estonia, June 11–12, 2018, Proceedings, Numero 336, 2018, Pagina/e 56-71, ISBN 978-3-030-04839-6
Editore: Springer International Publishing
DOI: 10.1007/978-3-030-04840-2_5

Width of Minima Reached by Stochastic Gradient Descent is Influenced by Learning Rate to Batch Size Ratio

Autori: Stanislaw Jastrzębski, Zachary Kenton, Devansh Arpit, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos Storkey
Pubblicato in: Artificial Neural Networks and Machine Learning – ICANN 2018 - 27th International Conference on Artificial Neural Networks, Rhodes, Greece, October 4-7, 2018, Proceedings, Part III, Numero 11141, 2018, Pagina/e 392-402, ISBN 978-3-030-01423-0
Editore: Springer International Publishing
DOI: 10.1007/978-3-030-01424-7_39

Augmenting Image Classifiers Using Data Augmentation Generative Adversarial Networks

Autori: Antreas Antoniou, Amos Storkey, Harrison Edwards
Pubblicato in: Artificial Neural Networks and Machine Learning – ICANN 2018 - 27th International Conference on Artificial Neural Networks, Rhodes, Greece, October 4-7, 2018, Proceedings, Part III, Numero 11141, 2018, Pagina/e 594-603, ISBN 978-3-030-01423-0
Editore: Springer International Publishing
DOI: 10.1007/978-3-030-01424-7_58

Towards Secure Collaborative AI Service Chains

Autori: Vida Ahmadi Mehri
Pubblicato in: 2019
Editore: Blekinge Institute of Technology

Robustness to adversarial examples can be improved with overfitting

Autori: Oscar Deniz, Anibal Pedraza, Noelia Vallez, Jesus Salido, Gloria Bueno
Pubblicato in: International Journal of Machine Learning and Cybernetics, Numero 11/4, 2020, Pagina/e 935-944, ISSN 1868-8071
Editore: Springer Science + Business Media
DOI: 10.1007/s13042-020-01097-4

È in corso la ricerca di dati su OpenAIRE...

Risultati finali

Pubblicazioni

Condividi questa pagina

Scarica