– One Paper is accepted by the 58th Design Automation Conference (DAC) 2021.
“Helios: Heterogeneity-Aware Federated Learning with Dynamically Balanced Collaboration”

Intelligence Fusion Laboratory
Electrical and Computer Engineering | George Mason University & University of Maryland, Baltimore County
News about the lab’s paper publication.
– One Paper is accepted by the 58th Design Automation Conference (DAC) 2021.
“Helios: Heterogeneity-Aware Federated Learning with Dynamically Balanced Collaboration”
– Two Papers are accepted by the 5th ACM/IEEE Symposium on Edge Computing (SEC) 2020.
“Exploring the Design Space of Efficient Deep Neural Networks”
“Exploring Decentralized Collaboration in Heterogeneous Edge Training”
– One Paper is accepted by the 33rd IEEE Conference on Computer Vision and Pattern Recognition, Workshop on Vision for all Seasons: Adverse Weather and Lighting Conditions (ECCV) 2020.
“An Image Enhancing Pattern-based Sparsity for Real-time Inference on Mobile Devices”
– One Paper is accepted by the 33rd IEEE Conference on Computer Vision and Pattern Recognition, Workshop on Vision for all Seasons: Adverse Weather and Lighting Conditions (CVPR-V4AS) 2020.
“Unsupervised Domain Adaptation for Object Detection via Cross-Domain Semi-Supervised Learning”
– “AntiDOte: Attention-based Dynamic Optimization for Neural Network Runtime Efficiency ” received the Best Paper Nomination from the 23rd International Conference on Design Automation and Test in Europe (DATE) 2020.
– Three Papers are accepted by the 24th Asia and South Pacific Design Automation Conference (ASP-DAC) 2019.
“LanCe: A Comprehensive and Lightweight CNN Defense Methodology against Physical Adversarial Attacks on Embedded Multimedia Applications”
“Tiny but Accurate: A Pruned, Quantized and Optimized Memristor Crossbar Framework for Ultra-Efficient DNN Implementation”
– Three Papers are accepted by the 23rd International Conference on Design Automation and Test in Europe (DATE) 2020.
“AntiDOte: Attention-based Dynamic Optimization for Neural Network Runtime Efficiency”
“DCCNN: Computational Flow Redefinition for Efficient CNN Inference through Model Structural Decoupling”
– One Paper is accepted by the 5th ACM/IEEE Symposium on Edge Computing (SEC) 2019.
“Task-Adaptive Incremental Learning for Intelligent Edge Devices”
– One Papers is accepted by the 28th ACM International Conference on Information and Knowledge Management (CIKM) 2019.
“Multi-stage Deep Classifier Cascades for Open World Recognition”
– One Papers is accepted by the 30th British Machine Vision Conference (BMVC) 2019.
“Functionality-Oriented Convolutional Filter Pruning”