NEWS

Our team won the first-place award at the 2024 University Demonstration at DAC, The Chips to Systems Conference for the project titled "HiRISE: High-Resolution Image Scaling for Edge ML via In-Sensor Compression and Selective ROI."  Congratulations to our PhD research assistants Brendan Reidy and Peyton Chandarana for their outstanding demonstration!

Our paper titled "Towards Efficient Deployment of Hybrid SNNs on Neuromorphic and Edge AI Hardware"  has been accepted for presentation at the 2024 ACM International Conference on Neuromorphic Systems (ICONS'24). Congratulations to Blake Seekings as the lead author and Peyton Chandarana, Mahsa Ardakani, and Mohammadreza Mohammadi as co-authors of this paper!

Our proposal titled "Cross-Layer Design Automation for In-Memory Analog Computing" has been awarded funding by the National Science Foundation (NSF) Software & Hardware Foundation (SHF) program. 

Our project titled "HiRISE: High-Resolution Image Scaling for Edge ML via In-Sensor Compression and Selective ROI" has been accepted for the University Demo at DAC 2024! Congratulations to Brendan Reidy for leading this project and Peyton Chandarana for his supporting role!

Our paper titled "Multi-Objective Neural Architecture Search for In-Memory Computing"  has been accepted for presentation at the 2024 IEEE Computer Society Annual Symposium on VLSI (ISVLSI'24). Congratulations to Md Hasibul Amin as the lead author and  Mohammadreza Mohammadi as a co-author of this paper!

Our research on "Flex-TPU: A Flexible TPU Architecture with Runtime Reconfigurable Dataflow"  has been awarded 1st Place at the Spring 2024 CSE Symposium. Congratulations to the iCAS Lab's members involved in this project including Mohammed Elbtity and Peyton Chandarana!

Our NSF Career proposal entitled "Heterogeneous Neuromorphic and Edge Computing Systems for Realtime Machine Learning Technologies" was awarded funding by the National Science Foundation. Congratulations to the iCAS Lab as a whole for helping with this monumental step in research!

Congratulations to Mohammed Elbtity for being the 1st Ph.D. student in the iCAS Lab to successfully defend his dissertation entitled "Approximate Computing and In-Memory Computing: The best of the two worlds!"

We have two papers accepted for Oral presentation and one paper for Poster presentation at the 2024 Design Automation Conference (DAC):

Oral Presentation 1:  "HiRISE: High-Resolution Image Scaling for Edge ML via In-Sensor Compression and Selective ROI," Congratulations to Brendan Reidy as the lead author and  Mohammadreza Mohammadi as a co-author of this paper. 

Oral Presentation 2: "Lightator: An Optical Near-Sensor Accelerator with Compressive Acquisition Enabling Versatile Image Processing." Congratulations to Brendan Reidy as a co-author of this paper! The ACAD Lab led this work at the New Jersey Institute of Technology.

Poster Presentation: "PixelPrune: Sparse Object Detection for AIoT Systems via In-Sensor Segmentation and Adaptive Data Transfer." Congratulations to Mohammadreza Mohammadi as the lead author and Brendan Reidy as a co-author of this paper!

Our paper "Realtime Facial Expression Recognition: Neuromorphic Hardware vs. Edge AI Accelerators" was presented at ICMLA 2023. Congratulations to Heath Smith, Blake Seekings, and Mohammadreza Mohammadi as co-authors of this paper!

Our paper "Caveline Detection at the Edge for Autonomous Underwater Cave Exploration and Mapping," was presented at ICMLA 2023. Congratulations to Mohammadreza Mohammadi as the lead author of this paper!

Our research on "Facial Expression Recognition at the Edge: CPU vs GPU vs VPU vs TPU"  has been awarded 1st Place at the Spring 2023 CSE Symposium. Congratulations to the iCAS Lab's members involved in this project including Mohammadreza Mohammadi, Heath Smith, Lareb Khan, and Mahsa Majdzadeh Ardakani.

Our poster "Facial Expression Recognition at the Edge: CPU vs GPU vs VPU vs TPU"  has been awarded 1st Place at the Engineering and Computing Track of Discover USC undergraduate poster presentations. Congratulations to iCAS Lab's undergraduate student Lareb Khan and her Ph.D. mentor Mohammadreza Mohammadi!

The iCAS Lab presented two papers at the 24th International Symposium on Quality Electronic Design (ISQED'23):
1)  "A Novel Stochastic LSTM Model Inspired by Quantum Machine Learning," presented by Joseph Lindsay.
2) "Application of Machine Learning for Quality Risk Factor Analysis of Electronic Assemblies," presented by Brendan Reidy.

The iCAS Lab has three papers accepted for oral presentation at the GLSVLSI'23:
1)  "Heterogeneous Integration of In-Memory Analog Computing Architectures with Tensor Processing Units,"  by Mohammed Elbtity, Brendan Reidy, Md Hasibul Amin.
2) "Facial Expression Recognition at the Edge: CPU vs GPU vs VPU vs TPU," by Mohammadreza Mohammadi, Heath Smith, and Lareb Khan.
3) "IMAC-Sim: A Circuit-level Simulator For In-Memory Analog Computing Architectures," by Md Hasibul Amin and Mohammed E. Elbtity.

Congratulations to all the authors and co-authors!

Our paper "Real-time Transformer Inference on Edge AI Accelerators," has been accepted for presentation at RTAS'23. Congratulations to Brendan Reidy, Mohammadreza Mohammadi, Mohammed Elbtity, and Heath Smith as co-authors of this paper.

The iCAS lab has received recognition as a recipient of the ZPRIZE competition's Open Division award for "Accelerating Number Theoretic Transform (NTT) Operations on an FPGA." Only two academic labs were amongst the awardees with iCAS being the only one based in the US.

Congratulations to iCAS's graduate students, Mohammed Elbtity and Joseph Lindsay, for their leadership in hardware and algorithm development, and Peyton Chandarana and Mohammadreza Mohammadi, for their supporting contributions to the project.

​Our paper "Xbar-partitioning: A Practical Way for Parasitics and Noise Tolerance in Analog IMC Circuits", has been accepted for publication on the special issue "Memristive Circuits and Systems for Edge-Computing Applications" at the IEEE Journal on Emerging and Selected Topics in Circuits and Systems (JETCAS)! Congratulations to Md Hasibul Amin and Mohammed Elbtity as co-authors!

​Our paper "Energy-Efficient Deployment of Machine Learning Workloads on Neuromorphic Hardware", has been accepted for presentation at the 2022 International Green and Sustainable Computing (IGSC) Conference! Congratulations to Peyton Chandarana, MohammadReza Mohammadi, and Blake Seekings as co-authors!

​Our paper "Static Hand Gesture Recognition for American Sign Language using Neuromorphic Hardware", has been accepted for publishing in the IOPScience: Neuromorphic Computing and Engineering! Congratulations to MohammadReza Mohammadi, Peyton Chandarana, Blake Seekings, and Sara Hendrix as co-authors!

​Our paper "APTPU: Approximate Computing Based Tensor Processing Unit", has been accepted for publishing in IEEE Transaction on Circuits and Systems I (TCAS-I)! Congratulations to Mohammed Elbtity, Peyton Chandarana, and Brendan Reidy as co-authors of this paper in this prestigious journal!

Our poster "Static American Sign Language Recognition Using Neuromorphic Hardware" was awarded 1st Place at the Engineering and Computing Track of Discover UofSC undergraduate poster presentations. Congratulations to iCAS Lab's undergraduate students Blake Seekings and Sara Hendrix with their Ph.D. mentors MohammadReza and Peyton Chandarana!

​Our paper "MRAM-based Analog Sigmoid Function for In-memory Computing", has been accepted for presentation at the 2022 Great Lakes Symposium on VLSI (GLSVLSI '22), to be held in Irvine, CA, USA from June 6 through June 8, 2022.
Congratulations to Md Hasibul Amin, Mohammed Elbtity, and Mohammadreza Mohammadi as co-authors of this paper!

Sara Hendrix received the Magellan Journey undergraduate research grant in the amount of $1000.

​Our paper "Interconnect Parasitics and Partitioning in Fully-Analog In-Memory Computing Architectures", has been accepted for presentation at the 2022 IEEE International Symposium on Circuits & Systems (ISCAS '22), to be held in Austin, Texas USA from May 28 through June 1, 2022. Congratulations to Md Hasibul Amin and Mohammed Elbtity as co-authors of this paper!

Our proposal "Attention Tracking for Intuitive Robot Tutors using Neuromorphic Computing" has been awarded funding through the Magellan Scholar Program in the amount of $2500. Congratulations to Blake Seekings and Peyton Chandarana for their work on this proposal!

​Our paper "An Adaptive Sampling and Edge Detection Approach for Encoding Static Images for Spiking Neural Networks" has been accepted to be presented at the  2021 International Green and Sustainable Computing Conference (IGSC) in the workshop on Neuromorphic Computing: from Material to Algorithm (NeuMA). This paper was the result of the course CSCE 790: Topics in Computing course on Neuromorphic Computing. Congratulations to Peyton Chandarana and Junlin Ou as co-authors of this paper!

Congratulations to Mohammed E. Elbtity for being selected to participate in the Design Automation Conference (DAC) Young Professionals Program!

Our poster "Approximate Computing Based Tensor Processing Unit For High-Performance AI Compute" has been accepted to be presented at the 2021 IBM-IEEE AI Compute Symposium (AICS). Congratulations to Mohammed E. Elbtity, Peyton Chandarana, and Brendan Reidy as co-authors of this paper! 

Our paper "An In-Memory Analog Computing Co-Processor for Energy-Efficient CNN Inference on Mobile Devices" has been selected as the best paper of the ISVLSI 2021 conference. This is a collaborative work with Lehigh University on the In-Memory Analog Computing project. Congratulations to Mohammed Elbtity and Brendan Reidy as co-authors of this paper! 

Congratulations to Mohammed E. Elbtity for being selected to receive the ISVLSI 2021 NSF student travel grant award!

​Our paper "An In-Memory Analog Computing Co-Processor for Energy-Efficient CNN Inference on Mobile Devices" has been accepted to be presented at the  2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI). This is a collaborative work with Lehigh University on the In-Memory Analog Computing project. Congratulations to Mohammed Elbtity and Brendan Reidy as co-authors of this paper!

​Our proposal "An Energy-Efficient Neuromorphic Patient Monitoring System on Wearable Devices" has been awarded the ASPIRE grant from the Office of the Vice President for Research at the University of South Carolina.

​Congratulations to Blake Seekings for being selected for the Summer 2021 McNAIR Junior Fellows (MJF) Program of the UofSC College of Engineering and Computing for his work on Neuro-Edge: Neuromorphic-Enhanced Edge Computing!

​Congratulations to Mohammed E. Elbtity for being selected for an INFOCOM 2021 NSF Student Conference Award!

Our collaborative paper on the Deep-Morphology project titled "Application of Machine Learning in Recognition and Analysis of TSV Extrusion Profiles with Multiple Morphology," has been accepted to be presented at the IEEE 71st  Electronic Components and Technology Conference (ECTC) conference. Congratulations to Joseph Lindsay, Brendan Reidy, and Will  Duggan as co-authors of this paper!

​We are happy to announce that iCAS Lab has started a collaboration with Juniper Networks, Inc., on the topic of “Machine Learning-Based Data Analytics Tools for Advanced Quality Management of Electronic Circuits”.

 Our proposal “Real-time Patient Monitoring on Wearable Devices Using Neuromorphic Computing” has been awarded funding through the Magellan Scholar Program in the amount of $2500. Congratulations to Praful Chunchu and Denise Davis for their work on this proposal!

​Our paper “A Single-Cycle MLP Classifier Using Analog MRAM-based Neurons and Synapses”, has been accepted to be presented at the 2020 International Conference on Rebooting Computing (ICRC).

​Our paper “An In-Memory Analog Computing Co-Processor for Deep Learning at the Edge”, has been accepted to be presented at the 2020 IBM IEEE AI Compute Symposium.

​Our paper “TSV Extrusion Morphology Classification Using Deep Convolutional Neural Networks”, has been accepted to be presented at the 19th IEEE International Conference on Machine Learning and Applications (ICMLA).