The paper titled “DualPIM: A Dual-Precision and Low-Power CNN Inference Engine Using SRAM- and eDRAM-based Processing-in-Memory Arrays” is accepted to IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS). Congrats, Sangwoo, Jaehyun and Huiseoung! This is a collaborative research with Prof. Jong-Hyeok Yoon at DGIST.

The paper titled “Implication of Optimizing NPU Dataflows on Neural ArchitectureSearch for Mobile Devices” is accepted to ACM Transactions on Design Automation of Electronic Systems (TODAES). Congrats, Jooyeon, Junsang and Seunghyun!

The paper titled "ZeBRA: Precisely Destroying Neural Networks with Zero-Data Based Repeated Bit Flip Attack" is accepted to British Machine Vision Conference (BMVC). Congrats, Dahoon Park! This work is a collaborated research between DGIST and Hongik University (Prof. Kon-Woo Kwon). 

The paper titled "AutoRelax: HW-SW Co-Optimization for Efficient SpGEMM Operations with Automated Relaxation in Deep Learning" is accepted to IEEE Transactions on Emerging Topics in Computing (TETC). Congrats, Sehun Park!

The paper titled "Deep Partitioned Training from Near-Storage Computing to DNN Accelerators" is accepted to IEEE Computer Architecture Letters (CAL). Congrats, Yongjoo Jang!

[2021.04.04] Our collaborated research paper with POSTECH titled "Design and Analysis of Approximate Compressors for Balanced Error Accumulation in MAC Operator" is accepted to IEEE TCAS-I. Congrats, Gunho Park! [2021.04.14] Our collaborated research paper with POSTECH titled "High-Throughput Near-Memory Processing on CNNs with 3D HBM-like Memory " is accepted to ACM TODAES. Congrats, Naebeom Park!