Automated and efficient Bangla signboard detection, text extraction, and novel categorization method for underrepresented languages in smart cities by Tanmoy Mazumder, Fariha Nusrat, Abu Bakar Siddique Mahi, Jolekha Begum Brishty, Rashik Rahman, Tanjina Helaly

Get full text

Randomized Quantization for Privacy in Resource Constrained Machine Learning at-the-Edge and Federated Learning by Ce Feng, Parv Venkitasubramaniam

“…Through rigorous theoretical analysis and extensive experiments on benchmark datasets, we demonstrate that these methods significantly enhance the utility-privacy trade-off and computational efficiency in both ML-at-the-edge and FL systems. RQP-SGD is evaluated on MNIST and the Breast Cancer Diagnostic dataset, showing an average 10.62% utility improvement over the deterministic quantization-based projected DP-SGD while maintaining (1.0, 0)-DP. …”
Get full text

Lost-minimum post-training parameter quantization method for convolutional neural network by Fan ZHANG, Yun HUANG, Zizhuo FANG, Wei GUO

“…To solve the problem that that no dataset is available for model quantization in data-sensitive scenarios, a model quantization method without using data sets was proposed.Firstly, according to the parameters of batch normalized layer and the distribution characteristics of image data, the simulated input data was obtained by error minimization method.Then, by studying the characteristics of data rounding, a factor dynamic rounding method based on loss minimization was proposed.Through quantitative experiments on classification models such as GhostNet and target detection models such as M2Det, the effectiveness of the proposed quantification method for image classification and target detection models was verified.The experimental results show that the proposed quantization method can reduce the model size by about 75%, effectively reduce the power loss and improve the computing efficiency while basically maintaining the accuracy of the original model.…”
Get full text

AFQSeg: An Adaptive Feature Quantization Network for Instance-Level Surface Crack Segmentation by Shaoliang Fang, Lu Lu, Zhu Lin, Zhanyu Yang, Shaosheng Wang

“…To address these issues, this paper proposes a crack detection model based on adaptive feature quantization, which primarily consists of a maximum soft pooling module, an adaptive crack feature quantization module, and a trainable crack post-processing module. …”
Get full text

NeuBridge: bridging quantized activations and spiking neurons for ANN-SNN conversion by Yuchen Yang, Jingcheng Liu, Chengting Yu, Chengyi Yang, Gaoang Wang, Aili Wang

“…Spiking neural networks (SNNs) offer a promising avenue for energy-efficient computations on neuromorphic hardware, leveraging the unique advantages of spike-based signaling. …”
Get full text

Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication by Yunshan Lv, Hailing Xiong, Fuqing Zhang, Shengying Dong, Xiangguang Dai

“…To solve the problem over a communication-efficient manner, we propose a novel quantized decentralized adaptive momentum gradient descent algorithm based on the adaptive momentum estimation methods, where quantified information is exchanged between agents. …”
Get full text

Convolution Smooth: A Post-Training Quantization Method for Convolutional Neural Networks by Yongyuan Chen, Zhendao Wang

“…Convolutional neural network (CNN) quantization is an efficient model compression technique primarily used for accelerating inference and optimizing resources. …”
Get full text

Hierarchical Mixed-Precision Post-Training Quantization for SAR Ship Detection Networks by Hang Wei, Zulin Wang, Yuanhan Ni

“…However, limited satellite platform resources present a significant challenge. Post-training quantization (PTQ) provides an efficient method for pre-training neural networks to effectively reduce memory and computational resources without retraining. …”
Get full text

Enriched HARQ Feedback for Link Adaptation in 6G: Optimizing Uplink Overhead for Enhanced Downlink Spectral Efficiency by Aritra Mazumdar, Abolfazl Amiri, Stefano Paris, Klaus I. Pedersen, Ramoni Adeogun

“…First, our learning-driven adaptive quantization (LAQ) employs a-priori statistics to refine delta MCS quantization within fixed-size UE feedback. …”
Get full text

FL-QNNs: Memory Efficient and Privacy Preserving Framework for Peripheral Blood Cell Classification by Meenakshi Aggarwal, Vikas Khullar, Nitin Goyal, Bhavani Sankar Panda, Hardik Doshi, Nafeesh Ahmad, Vivek Bhardwaj, Gaurav Sharma

“…This study proposes a resource efficient, privacy preserving, optimized memory framework by incorporating two approaches: Federated learning and quantized neural network (FL-QNNs) for peripheral blood cell (PBC) image classification. …”
Get full text

Research of channel quantization and feedback strategies based on multiuser diversity MIMO-OFDM systems by LIANG Xue-jun, ZHU Guang-xi, SU Gang, WANG De-sheng

“…Firstly, a quantization method was proposed by quantized value indicating the modulation level instead of the full values of channel quality information（CQI） and the achievable average spectrum efficiency showed no loss compared with perfect case.Secondly, employment of the integrated design that combined with opportunistic, best, and hybrid feedback scheme was considered and the close-form expression of average spectrum efficiency was deduced in various case.Finally, the calculation of optimal feedback parameters was confirmed from two aspects of feedback channel capacity and capacity relative loss.Extensive simulations were presented to evaluate these proposed strategies.The results match with the numeral analysis very well.The proposed partial feedback schemes can reduce the feedback load greatly with the same system capability, only if the feedback parameters be chosen properly.Wherein, the hybrid feedback combined with quantization performs best and provides the instruction to design the channel feedback of practical systems.…”
Get full text

Smoothed per-tensor weight quantization: a robust solution for neural network deployment by Xin Chang

“…This paper introduces a novel method to improve quantization outcomes for per-tensor weight quantization, focusing on enhancing computational efficiency and compatibility with resource-constrained hardware. …”
Get full text

Fully Quantized Matrix Arithmetic-Only BERT Model and Its FPGA-Based Accelerator by Hiroshi Fuketa, Toshihiro Katashita, Yohei Hori, Masakazu Hioki

“…In this paper, we propose a fully quantized matrix arithmetic-only BERT (FQ MA-BERT) model to enable efficient natural language processing. …”
Get full text

Reducing Memory and Computational Cost for Deep Neural Network Training with Quantized Parameter Updates by Leo Buron, Andreas Erbslöh, Gregor Schiele

“…For embedded devices, both memory and computational efficiency are essential due to their constrained resources. …”
Get full text

Optimizing Deep Learning Models for Resource‐Constrained Environments With Cluster‐Quantized Knowledge Distillation by Niaz Ashraf Khan, A. M. Saadman Rafat

“…To address these issues, we propose Cluster‐Quantized Knowledge Distillation (CQKD), a novel framework that integrates structured pruning with knowledge distillation, incorporating cluster‐based weight quantization directly into the training loop. …”
Get full text

Study of algorithmic approaches to digital signal filtering and the influence of input quantization on output accuracy by Shermuradova Malika, Gadoeva Mavlyuda, Rahmatov Shukhrat, Abdullaev Uktamjon, Aralov Dilshod

“…The research supports the broader integration of AI-driven technologies in modern automation systems, paving the way for more adaptive, efficient, and fault-tolerant control mechanisms in complex environments.…”
Get full text

FPGA Acceleration With Hessian-Based Comprehensive Intra-Layer Mixed-Precision Quantization for Transformer Models by Woohong Byun, Jongseok Woo, Saibal Mukhopadhyay

“…Our concurrent quantization method balances the benefits of row-wise weight quantization and Query-Key coupled activation quantization while maximizing energy efficiency through multi-precision optimization. …”
Get full text

GENERATING OF OPTIMAL QUANTIZATION LEVELS OF CONTROL CURRENTS FOR LINEAR STEPPING DRIVES OF PRECISION MOTION SYSTEMS by I. V. Dainiak, V. P. Oger, D. S. Tsitko, D. G. Begun

“…The investigations have proved an efficiency of the proposed algorithm and methodology for forming coordinate discrete grid.…”
Get full text

QDLTrans: Enhancing English Neural Machine Translation With Quantized Attention Block and Tunable Dual Learning by Xing Liu

“…Specifically, our method outperforms existing approaches in both translation quality and efficiency, offering a balanced solution for enhancing NMT in low-resource settings.…”
Get full text

Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering by Thommas K. S. Flores, Morsinaldo Medeiros, Marianne Silva, Daniel G. Costa, Ivanovitch Silva

“…This study introduces a novel method to optimize Post-Training Quantization (PTQ), a widely used technique for reducing model size, by integrating Vector Quantization (VQ) with incremental clustering. …”
Get full text