Chen, Wei; Bhardwaj, Kartikeya; Marculescu, Radu FedMAX: Mitigating Activation Divergence for Accurate and Communication-Efficient Federated Learning (Journal Article) The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD) , 2020. (Links) @article{chen2020fedmax,
title = {FedMAX: Mitigating Activation Divergence for Accurate and Communication-Efficient Federated Learning},
author = {Wei Chen and Kartikeya Bhardwaj and Radu Marculescu},
url = {https://arxiv.org/abs/2004.03657},
year = {2020},
date = {2020-01-01},
journal = {The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD) },
keywords = {},
pubstate = {published},
tppubtype = {article}
}
|
Farcas, Allen-Jasmin; Li, Guihong; Bhardwaj, Kartikeya; Marculescu, Radu A Hardware Prototype Targeting Distributed Deep Learning for On-Device Inference (Inproceedings) Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, pp. 398–399, 2020. (Links) @inproceedings{farcas2020hardware,
title = {A Hardware Prototype Targeting Distributed Deep Learning for On-Device Inference},
author = {Allen-Jasmin Farcas and Guihong Li and Kartikeya Bhardwaj and Radu Marculescu},
url = {https://openaccess.thecvf.com/content_CVPRW_2020/html/w28/Farcas_A_Hardware_Prototype_Targeting_Distributed_Deep_Learning_for_On-Device_Inference_CVPRW_2020_paper.html},
year = {2020},
date = {2020-01-01},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
pages = {398--399},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
|
Lin, Ching-Yi; Marculescu, Radu Model Personalization for Human Activity Recognition (Inproceedings) 2020 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), pp. 1–7, IEEE 2020. (Links) @inproceedings{lin2020model,
title = {Model Personalization for Human Activity Recognition},
author = {Ching-Yi Lin and Radu Marculescu},
url = {https://ieeexplore.ieee.org/abstract/document/9156229},
year = {2020},
date = {2020-01-01},
booktitle = {2020 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)},
pages = {1--7},
organization = {IEEE},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
|
Marculescu, Radu; Marculescu, Diana; Ogras, Umit Edge AI: Systems Design and ML for IoT Data Analytics (Inproceedings) Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp. 3565–3566, 2020. (Links) @inproceedings{marculescu2020edge,
title = {Edge AI: Systems Design and ML for IoT Data Analytics},
author = {Radu Marculescu and Diana Marculescu and Umit Ogras},
url = {https://sites.google.com/utexas.edu/edgeaitutorial/home?authuser=0},
year = {2020},
date = {2020-01-01},
booktitle = {Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining},
pages = {3565--3566},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
|
Bhardwaj, Kartikeya; Chen, Wei; Marculescu, Radu New Directions in Distributed Deep Learning: Bringing the Network at Forefront of IoT Design (Journal Article) arXiv preprint arXiv:2008.10805, 2020. (Links) @article{bhardwaj2020new,
title = {New Directions in Distributed Deep Learning: Bringing the Network at Forefront of IoT Design},
author = {Kartikeya Bhardwaj and Wei Chen and Radu Marculescu},
url = {https://arxiv.org/pdf/2008.10805},
year = {2020},
date = {2020-01-01},
journal = {arXiv preprint arXiv:2008.10805},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
|
Bhardwaj, Kartikeya; Suda, Naveen; Marculescu, Radu Dream distillation: A data-independent model compression framework (Journal Article) arXiv preprint arXiv:1905.07072, 2019. (Links) @article{bhardwaj2019dream,
title = {Dream distillation: A data-independent model compression framework},
author = {Kartikeya Bhardwaj and Naveen Suda and Radu Marculescu},
url = {https://arxiv.org/abs/1905.07072},
year = {2019},
date = {2019-01-01},
journal = {arXiv preprint arXiv:1905.07072},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
|
Internet of Things
Internet of Things (IoT) represents a paradigm shift from the traditional Internet and Cloud computing to a new reality where all “things” are connected to the Internet. Indeed, it has been estimated that the number of connected IoT-devices will reach one trillion by 2035. Such an explosive growth in IoT-devices necessitates new breakthroughs in AI research that can help efficiently deploy intelligence at the edge. Given that IoT-devices are extremely resource-constrained (e.g., small memory, low operating frequencies for energy efficiency), we focus primarily on challenges related to enabling deeplearning models at the edge.
Model Compression & Optimization
Model compression has emerged as an important area of research for deploying deep learning models on IoT devices. However, model compression is not a sufficient solution to fit the models within the memory of a single device; as a result we need to distribute them across multiple devices. This leads to a distributed inference paradigm in which communication costs represent another major bottleneck. To this end, we focus on knowledge distillation and ‘teacher’ – ‘student’ type of architectures for distributed model compression, as well as data independent model compression.
Federated Learning
Large amounts of data are generated nowadays on edge devices, such as phones, tablets, and wearable devices. However, since data on such personal devices is highly sensitive, training ML models by sending the users’ local data to a centralized server clearly involves significant privacy risks. Hence, in order to enable intelligence for these privacy-critical applications, Federated Learning (FL) has become the de facto paradigm for training ML models on local devices without sending data to the cloud. Our research in this direction focuses on developing new FL approaches that exploit data and devices heterogeneity.