Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Diana Marculescu

Width Transfer: On the (In)variance of Width Optimization


Apr 24, 2021
Ting-Wu Chin, Diana Marculescu, Ari S. Morcos

* Full paper accepted at CVPR Workshops 2021; a 4-page abridged version is accepted at ICLR 2021 NAS Workshop 

  Access Paper or Ask Questions

DeepNVM++: Cross-Layer Modeling and Optimization Framework of Non-Volatile Memories for Deep Learning


Dec 08, 2020
Ahmet Inci, Mehmet Meric Isgenc, Diana Marculescu

* 12 pages, 10 figures 

  Access Paper or Ask Questions

The Architectural Implications of Distributed Reinforcement Learning on CPU-GPU Systems


Dec 08, 2020
Ahmet Inci, Evgeny Bolotin, Yaosheng Fu, Gal Dalal, Shie Mannor, David Nellans, Diana Marculescu

* To appear in the proceedings of the 6th Workshop on Energy Efficient Machine Learning and Cognitive Computing (EMC2) 2020 

  Access Paper or Ask Questions

One Weight Bitwidth to Rule Them All


Aug 28, 2020
Ting-Wu Chin, Pierce I-Jen Chuang, Vikas Chandra, Diana Marculescu

* Accepted at ECCV 2020 Embedded Vision Workshop (Best paper) 

  Access Paper or Ask Questions

PareCO: Pareto-aware Channel Optimization for Slimmable Neural Networks


Jul 23, 2020
Ting-Wu Chin, Ari S. Morcos, Diana Marculescu

* preprint, 4-page abridged versions have been accepted at non-archival venues including RealML and DMMLSys workshops at ICML'20 and DLP-KDD and AdvML workshops at KDD'20 

  Access Paper or Ask Questions

Improving the Adversarial Robustness of Transfer Learning via Noisy Feature Distillation


Feb 07, 2020
Ting-Wu Chin, Cha Zhang, Diana Marculescu

* Preprint 

  Access Paper or Ask Questions

Single-Path Mobile AutoML: Efficient ConvNet Design and NAS Hyperparameter Optimization


Jul 01, 2019
Dimitrios Stamoulis, Ruizhou Ding, Di Wang, Dimitrios Lymberopoulos, Bodhi Priyantha, Jie Liu, Diana Marculescu

* Detailed extension (journal) of the Single-Path NAS ECMLPKDD'19 paper (arXiv:1904.02877) 

  Access Paper or Ask Questions

ViP: Virtual Pooling for Accelerating CNN-based Image Classification and Object Detection


Jun 19, 2019
Zhuo Chen, Jiyuan Zhang, Ruizhou Ding, Diana Marculescu


  Access Paper or Ask Questions

Single-Path NAS: Device-Aware Efficient ConvNet Design


May 10, 2019
Dimitrios Stamoulis, Ruizhou Ding, Di Wang, Dimitrios Lymberopoulos, Bodhi Priyantha, Jie Liu, Diana Marculescu

* ODML-CDNNR 2019 (ICML'19 workshop) oral presentation (extended abstract, required non-archival version). Full paper: arXiv:1904.02877 

  Access Paper or Ask Questions

LeGR: Filter Pruning via Learned Global Ranking


Apr 28, 2019
Ting-Wu Chin, Ruizhou Ding, Cha Zhang, Diana Marculescu

* 8 pages, 5 figures, 2 tables 

  Access Paper or Ask Questions

Single-Path NAS: Designing Hardware-Efficient ConvNets in less than 4 Hours


Apr 05, 2019
Dimitrios Stamoulis, Ruizhou Ding, Di Wang, Dimitrios Lymberopoulos, Bodhi Priyantha, Jie Liu, Diana Marculescu


  Access Paper or Ask Questions

FLightNNs: Lightweight Quantized Deep Neural Networks for Fast and Accurate Inference


Apr 05, 2019
Ruizhou Ding, Zeye Liu, Ting-Wu Chin, Diana Marculescu, R. D., Blanton


  Access Paper or Ask Questions

Regularizing Activation Distribution for Training Binarized Deep Networks


Apr 04, 2019
Ruizhou Ding, Ting-Wu Chin, Zeye Liu, Diana Marculescu


  Access Paper or Ask Questions

AdaScale: Towards Real-time Video Object Detection Using Adaptive Scaling


Feb 08, 2019
Ting-Wu Chin, Ruizhou Ding, Diana Marculescu

* Accepted to SysML 2019 (http://www.sysml.cc/) as oral contribution 

  Access Paper or Ask Questions

Understanding the Impact of Label Granularity on CNN-based Image Classification


Jan 21, 2019
Zhuo Chen, Ruizhou Ding, Ting-Wu Chin, Diana Marculescu

* 10 pages. Accepted at DSBDA with ICDM 

  Access Paper or Ask Questions

Learning-based Application-Agnostic 3D NoC Design for Heterogeneous Manycore Systems


Oct 20, 2018
Biresh Kumar Joardar, Ryan Gary Kim, Janardhan Rao Doppa, Partha Pratim Pande, Diana Marculescu, Radu Marculescu

* Under review at IEEE Transactions on Computers 

  Access Paper or Ask Questions

Layer-compensated Pruning for Resource-constrained Convolutional Neural Networks


Oct 18, 2018
Ting-Wu Chin, Cha Zhang, Diana Marculescu

* 11 pages, 8 figures, work in progress 

  Access Paper or Ask Questions

Hardware-Aware Machine Learning: Modeling and Optimization


Sep 14, 2018
Diana Marculescu, Dimitrios Stamoulis, Ermao Cai

* ICCAD'18 Invited Paper 

  Access Paper or Ask Questions

Designing Adaptive Neural Networks for Energy-Constrained Image Classification


Aug 07, 2018
Dimitrios Stamoulis, Ting-Wu Chin, Anand Krishnan Prakash, Haocheng Fang, Sribhuvan Sajja, Mitchell Bognar, Diana Marculescu

* This conference paper will appear in the proceedings of ICCAD 2018 

  Access Paper or Ask Questions

Tractable Learning and Inference for Large-Scale Probabilistic Boolean Networks


Jan 23, 2018
Ifigeneia Apostolopoulou, Diana Marculescu

* 16 pages 

  Access Paper or Ask Questions

HyperPower: Power- and Memory-Constrained Hyper-Parameter Optimization for Neural Networks


Dec 06, 2017
Dimitrios Stamoulis, Ermao Cai, Da-Cheng Juan, Diana Marculescu

* This conference paper will appear in the proceedings of DATE 2018 

  Access Paper or Ask Questions

LightNN: Filling the Gap between Conventional Deep Neural Networks and Binarized Networks


Dec 02, 2017
Ruizhou Ding, Zeye Liu, Rongye Shi, Diana Marculescu, R. D. Blanton


  Access Paper or Ask Questions

Machine Learning and Manycore Systems Design: A Serendipitous Symbiosis


Nov 30, 2017
Ryan Gary Kim, Janardhan Rao Doppa, Partha Pratim Pande, Diana Marculescu, Radu Marculescu

* To appear in a future publication of IEEE Computer 

  Access Paper or Ask Questions

NeuralPower: Predict and Deploy Energy-Efficient Convolutional Neural Networks


Oct 15, 2017
Ermao Cai, Da-Cheng Juan, Dimitrios Stamoulis, Diana Marculescu

* Accepted as a conference paper at ACML 2017 

  Access Paper or Ask Questions