Information, Inference and Machine Learning Group at University College London

Martin Ferianc

Fourth-year PhD Student

Contact

Where to Find Me

Malet Place Engineering Building
Torrington Place
London WC1E 7JE
United Kingdom

My Feed

Tweets by Martin

Research Interests

Hardware Optimisation and acceleration, Bayesian neural networks, Deep Learning

About

Martin Ferianc is a PhD candidate in the Department of Electronic and Electrical Engineering at University College London. His research interests include neural architecture search, Bayesian neural network, deep learning and hardware acceleration of neural networks. Martin has obtained an MEng in Electronic and Information Engineering from Imperial College London.

Publications

@InProceedings{10.1007/978-3-030-86365-4_39,
                  author="Ferianc, Martin
                  and Manocha, Divyansh
                  and Fan, Hongxiang
                  and Rodrigues, Miguel",
                  editor="Farka{\v{s}}, Igor
                  and Masulli, Paolo
                  and Otte, Sebastian
                  and Wermter, Stefan",
                  title="ComBiNet: Compact Convolutional Bayesian Neural Network for Image Segmentation",
                  booktitle="Artificial Neural Networks and Machine Learning -- ICANN 2021",
                  year="2021",
                  organization="Springer International Publishing",
                  address="Cham",
                  pages="483--494",
                  url = {https://arxiv.org/abs/2104.06957},
                  code = {https://github.com/martinferianc/ComBiNet},
                  isbn="978-3-030-86365-4"
                  }

@INPROCEEDINGS{9609847,
                      author={Ferianc, Martin and Que, Zhiqiang and Fan, Hongxiang and Luk, Wayne and Rodrigues, Miguel},
                      booktitle={2021 International Conference on Field-Programmable Technology (ICFPT)}, 
                      title={Optimizing {B}ayesian Recurrent Neural Networks on an {FPGA}-based Accelerator}, 
                      year={2021},
                      volume={},
                      number={},
                      pages={1-10},
                      organization={IEEE},
                      url = {https://arxiv.org/abs/2106.06048},
                      doi={10.1109/ICFPT52863.2021.9609847}}

@inproceedings{fan2018real,
                  title={A Real-Time Object Detection Accelerator with Compressed {SSDLite} on {FPGA}},
                  author={Fan, Hongxiang and Liu, Shuanglong and Ferianc, Martin and Ng, Ho-Cheung and Que, Zhiqiang and Liu, Shen and Niu, Xinyu and Luk, Wayne},
                  booktitle={2018 International Conference on Field-Programmable Technology (FPT)},
                  pages={14--21},
                  year={2018},
                  organization={IEEE},
                  url={https://ieeexplore.ieee.org/document/8742299}
                }
                @INPROCEEDINGS{DBLP:journals/corr/abs-2111-12787,
                  author={Hongxiang Fan and
                      Martin Ferianc and
                      Zhiqiang Que and
                      He Li and
                      Shuanglong Liu and
                      Xinyu Niu and
                      Wayne Luk},
                  booktitle={27th Asia and South Pacific Design Automation Conference
                      (ASP-DAC) 2022}, 
                  title={Algorithm and Hardware Co-design for Reconfigurable CNN Accelerator
                  }, 
                  year={2022},
                  volume={},
                  number={},
                  organization={IEEE},
                  pages={1-10},
                  url = {https://arxiv.org/abs/2111.12787}
              }
                
                @inproceedings{fan2019f,
                  title={{F-E3D}: {FPGA}-based acceleration of an efficient 3D convolutional neural network for human action recognition},
                  author={Fan, Hongxiang and Luo, Cheng and Zeng, Chenglong and Ferianc, Martin and Que, Zhiqiang and Liu, Shuanglong and Niu, Xinyu and Luk, Wayne},
                  booktitle={2019 IEEE 30th International Conference on Application-specific Systems, Architectures and Processors (ASAP)},
                  volume={2160},
                  pages={1--8},
                  year={2019},
                  organization={IEEE},
                  url={https://ieeexplore.ieee.org/abstract/document/8825127},
                  extra={Best Paper Award Nominee}
                }
                
                
                @inproceedings{fan2019static,
                  title={Static block floating-point quantization for convolutional neural networks on {FPGA}},
                  author={Fan, Hongxiang and Wang, Gang and Ferianc, Martin and Niu, Xinyu and Luk, Wayne},
                  booktitle={2019 International Conference on Field-Programmable Technology (ICFPT)},
                  pages={28--35},
                  year={2019},
                  organization={IEEE},
                  url={https://ieeexplore.ieee.org/abstract/document/8977902},
                  code={https://github.com/os-hxfan/Static_BFP_CNN},
                  extra={Best Paper Award Nominee}
                }
                
                @inproceedings{ferianc2020improving,
                  title={Improving Performance Estimation for {FPGA}-Based Accelerators for Convolutional Neural Networks},
                  author={Ferianc, Martin and Fan, Hongxiang and Chu, Ringo SW and Stano, Jakub and Luk, Wayne},
                  booktitle={International Symposium on Applied Reconfigurable Computing},
                  pages={3--13},
                  year={2020},
                  organization={Springer, Cham},
                  url = {https://link.springer.com/chapter/10.1007/978-3-030-44534-8_1},
                  code={https://github.com/martinferianc/Improving_Per_Esti_for_FPGA-based_Acc_for_CNNs-ARC2020},
                  media={https://www.ucl.ac.uk/electronic-electrical-engineering/news/2020/may/improving-performance-estimation-convolutional-neural-network-accelerators}
                }
                
                @misc{ferianc2021vinnas,
                  title={VINNAS: Variational Inference-based Neural Network Architecture Search}, 
                  author={Martin Ferianc and Hongxiang Fan and Miguel Rodrigues},
                  year={2021},
                  eprint={2007.06103},
                  archivePrefix={arXiv},
                  primaryClass={cs.LG},
                  url = {https://arxiv.org/abs/2007.06103}
              }
                @inproceedings{fan2020optimizing,
                  title={Optimizing {FPGA}-Based {CNN} Accelerator Using Differentiable Neural Architecture Search},
                  author={Fan, Hongxiang and Ferianc, Martin and Liu, Shuanglong and Que, Zhiqiang and Niu, Xinyu and Luk, Wayne},
                  booktitle={2020 IEEE 38th International Conference on Computer Design (ICCD)},
                  pages={465--468},
                  year={2020},
                  organization={IEEE},
                  url = {https://ieeexplore.ieee.org/document/9283571}
                }
                
                
                @article{liu2021toward,
                  title={Toward Full-Stack Acceleration of Deep Convolutional Neural Networks on {FPGA}s},
                  author={Liu, Shuanglong and Fan, Hongxiang and Ferianc, Martin and Niu, Xinyu and Shi, Huifeng and Luk, Wayne},
                  journal={IEEE Transactions on Neural Networks and Learning Systems},
                  year={2021},
                  publisher={IEEE},
                  url = {https://ieeexplore.ieee.org/iel7/5962385/6104215/09354062.pdf}
                }
                @article{ferianc2021improving,
                  title={Improving Performance Estimation for Design Space Exploration for Convolutional Neural Network Accelerators},
                  author={Ferianc, Martin and Fan, Hongxiang and Manocha, Divyansh and Zhou, Hongyu and Liu, Shuanglong and Niu, Xinyu and Luk, Wayne},
                  journal={Electronics},
                  volume={10},
                  number={4},
                  pages={520},
                  year={2021},
                  url={https://www.mdpi.com/2079-9292/10/4/520},
                  code={https://github.com/martinferianc/Improving_Per_Esti_for_FPGA-based_Acc_for_CNNs-ARC2020},
                  publisher={Multidisciplinary Digital Publishing Institute}
                }
                
                @misc{ferianc2021causal,
                  title={On Causal Inference for Data-free Structured Pruning}, 
                  author={Martin Ferianc and Anush Sankaran and Olivier Mastropietro and Ehsan Saboori and Quentin Cappart},
                  year={2021},
                  eprint={2112.10229},
                  booktitle={The AAAI-22 Workshop on Information-Theoretic Methods for Causal Inference and Discovery},
                  url = {https://arxiv.org/abs/2112.10229},
                  archivePrefix={arXiv},
                  primaryClass={cs.LG}
                }
                @InProceedings{9586137,
                  author={Fan, Hongxiang and Ferianc, Martin and Rodrigues, Miguel and Zhou, Hongyu and Niu, Xinyu and Luk, Wayne},
                  booktitle={2021 58th ACM/IEEE Design Automation Conference (DAC)}, 
                  title={High-Performance {FPGA}-based Accelerator for {B}ayesian Neural Networks}, 
                  year={2021},
                  volume={},
                  number={},
                  pages={1063-1068},
                  organization = {IEEE},
                  doi={10.1109/DAC18074.2021.9586137},
                  url = {https://arxiv.org/abs/2105.09163}
              }
                  @InProceedings{pmlr-v161-ferianc21a,
                      title = 	 {On the effects of quantisation on model uncertainty in {B}ayesian neural networks},
                      author =       {Ferianc, Martin and Maji, Partha and Mattina, Matthew and Rodrigues, Miguel},
                      booktitle = 	 {Proceedings of the Thirty-Seventh Conference on Uncertainty in Artificial Intelligence},
                      pages = 	 {929--938},
                      year = 	 {2021},
                      editor = 	 {de Campos, Cassio and Maathuis, Marloes H.},
                      volume = 	 {161},
                      series = 	 {Proceedings of Machine Learning Research},
                      month = 	 {27--30 Jul},
                      organization =    {PMLR},
                      pdf = 	 {https://proceedings.mlr.press/v161/ferianc21a/ferianc21a.pdf},
                      url = 	 {https://proceedings.mlr.press/v161/ferianc21a.html},
                      code = {https://github.com/martinferianc/quantised-bayesian-nets}
                  }

• [view online ] ,[view code ]

, ,

, Media Coverage

Martin Ferianc

Fourth-year PhD Student

Contact

Where to Find Me

My Feed

Research Interests

About

Publications

Refereed Articles

Books

Conference and Workshop Papers

Pre-prints and Other Publications