Toward Generating Quality Test Questions and Answers Using Quantized Low-Rank Adapters in LLMs by Jebum Choi, Seongjun Hong, Seoyoon Hong, Jiyeon Park, Eun-Sung Jung

Get full text

SparseBatch: Communication-efficient Federated Learning with Partially Homomorphic Encryption by Chong Wang, Jing Wang, Zheng Lou, Linghai Kong, WeiSong Tao, Yun Wang

Get full text

Learning from low precision samples by Ji In Choi, Madeleine Georges, Jung Ah Shin, Olivia Wang, Tiffany Zhu, Tapan Shah

Subjects: “…efficient machine learning…”
Get full text

Toward resource-efficient UAV systems: Deep learning model compression for onboard-ready weed detection in UAV imagery by Alwaseela Abdalla, Masara M.A. Mohammed, Oluwatola Adedeji, Peter Dotray, Wenxuan Guo

“…We fine-tuned the pruned model on the UAV dataset to mitigate any performance loss resulting from pruning. We then applied quantization to reduce the precision of numerical parameters and improve computational efficiency. …”
Get full text

Large language models for PHM: a review of optimization techniques and applications by Tingyi Yu, Junya Tang, Qingyun Yu, Li Li, Ying Liu, Raul Poler

Subjects: Get full text

Qptimization design of video encoder quantizer for general DSPs by GAN Yong1, SU Shi-mei2, ZHOU Bing2, QIAN De-pei 1

“…A new scheme of quantization which was constructed by an optimal quantization matrix was proposed.It was inefficient for the traditional quantization of MPEG video encoders on the general DSPs.Considering the traits of a DSP,which could realized one-step quantization by substituting shift for division.Meanwhile,this scheme used different quantization strategies according to the differences of the types of video blocks to be coded,and avoids the loss of the quality of images with the unified quantization.Experiments had shown that this scheme could obviously improve the efficiency of quantization,and subjective and objective quality of images coded as well.…”
Get full text

Efficient human activity recognition on edge devices using DeepConv LSTM architectures by Haotian Zhou, Xiujun Zhang, Yu Feng, Tongda Zhang, Lijuan Xiong

“…The device’s memory usage was 29.1 KB, flash usage was 189.6 KB, and the model’s average inference time was 21 milliseconds, requiring approximately 0.01395 GOP, with a computational performance of around 0.664 GOPS. Even after quantization, the model maintained an accuracy of 97% and an F1 score of 97%, ensuring efficient utilization of computational resources. …”
Get full text

Fully Quantized Neural Networks for Audio Source Separation by Elad Cohen, Hai Victor Habi, Reuven Peretz, Arnon Netzer

“…In this work, we focus on quantization, a leading approach for addressing these challenges. …”
Get full text

Quantization-based chained privacy-preserving federated learning by Ya Liu, Shumin Wu, Yibo Li, Fengyu Zhao, Yanli Ren

“…However, traditional FL schemes face significant challenges regarding communication efficiency, computational costs, and privacy preservation. …”
Get full text

Quantized convolutional neural networks: a hardware perspective by Li Zhang, Olga Krestinskaya, Mohammed E. Fouda, Ahmed M. Eltawil, Khaled Nabil Salama

“…Consistently, dedicated hardware accelerators are developed to further boost the execution efficiency of DNN models. In this work, we focus on Convolutional Neural Network (CNN) as an example of DNNs and conduct a comprehensive survey on various quantization and quantized training methods. …”
Get full text

Lightweight Road Environment Segmentation using Vector Quantization by J. Kwag, A. Yilmaz, C. Toth

“…In this work, we combined vector quantization with the lightweight image segmentation model MobileUNETR and used it as a baseline model for comparison to demonstrate its efficiency. …”
Get full text

A Novel Mixed-Precision Quantization Approach for CNNs by Dan Wu, Yanzhi Wang, Yuqi Fei, Guowang Gao

“…Mixed-precision quantization assigns different quantization precision to different layers of a CNN. …”
Get full text

Deep learning model for precise and rapid prediction of tomato maturity based on image recognition by Muhammad Waseem, Muhammad Muzzammil Sajjad, Laraib Haider Naqvi, Yaqoob Majeed, Tanzeel Ur Rehman, Tayyaba Nadeem

Subjects: “…Efficient net…”
Get full text

Hardware-Efficient Neural Network Digital Predistortion for Terahertz Power Amplifiers Using DeepShift and Pruning by Taishi Watanabe, Takeo Ohseki, Issei Kanno

Get full text

Bridging the Gap Between Computational Efficiency and Segmentation Fidelity in Object-Based Image Analysis by Fernanda Pereira Leite Aguiar, Irenilza de Alencar Nääs, Marcelo Tsuguio Okano

Subjects: Get full text

FPGA-QNN: Quantized Neural Network Hardware Acceleration on FPGAs by Mustafa Tasci, Ayhan Istanbullu, Vedat Tumen, Selahattin Kosunalp

“…The FPGA-QNN framework comes up with 12 accelerators based on multi-layer perceptron (MLP) and LeNet CNN models, each of which is associated with a specific combination of quantization and folding. The outputs from the performance evaluations on Xilinx PYNQ Z1 development board proved the superiority of FPGA-QNN in terms of resource utilization and energy efficiency in comparison to several recent approaches. …”
Get full text

THE ANALYSIS OF THE NOISE OF QUANTIZATION OF LINEAR STATIONARY FILTERS OF IMAGE PROCESSING by Nikolay Ivanovich Chervyakov, Pavel Alekseevich Lyakhov, Nikolay Nikolaevich Nagornov

“…The problem of quantization of the coefficients of arbitrary linear stationary filters in order to minimize the noise of this phenomenon and efficient hardware implementation of digital image processing methods is investigated in the paper. …”
Get full text

COMQ: A Backpropagation-Free Algorithm for Post-Training Quantization by Aozhong Zhang, Zi Yang, Naigang Wang, Yingyong Qi, Jack Xin, Xin Li, Penghang Yin

“…Post-training quantization (PTQ) has emerged as a practical approach to compress large neural networks, making them highly efficient for deployment. …”
Get full text

Quantized Feedback Control of Active Suspension Systems Based on Event Trigger by Jinwei Sun, Jingyu Cong, Weihua Zhao, Yonghui Zhang

“…In addition, the trigger mechanism can improve the working efficiency of actuators effectively.…”
Get full text

Optimizing binary neural network quantization for fixed pattern noise robustness by Francisco Javier Andreo-Oliver, Gines Domenech-Asensi, Jose Angel Diaz-Madrid, Ramon Ruiz-Merino, Juan Zapata-Perez

“…Abstract This work presents a comprehensive analysis of how extreme data quantization and fixed pattern noise (FPN) from CMOS imagers affect the performance of deep neural networks for image recognition tasks. …”
Get full text