0% Complete
Home
/
13th International Conference on Computer and Knowledge Engineering
Compressing Deep Neural Networks Using Explainable AI
Authors :
Kimia Soroush
1
Mohsen Raji
2
Behnam Ghavami
3
1- Shiraz university
2- Shiraz university
3- Shahid Bahonar University of Kerman
Keywords :
Deep Neural Networks،Compression،Explainable-AI
Abstract :
Abstract— Deep neural networks (DNNs) have demonstrated remarkable performance in many tasks but it often comes at a high computational cost and memory usage. Compression techniques, such as pruning and quantization, are applied to reduce the memory footprint of DNNs and make it possible to accommodate them on resource-constrained edge devices. Recently, explainable artificial intelligence (XAI) methods have been introduced with the purpose of understanding and explaining AI methods. XAI can be utilized to get to know the inner functioning of DNNs, such as the importance of different neurons and features in the overall performance of DNNs. In this paper, a novel DNN compression approach using XAI is proposed to efficiently reduce the DNN model size with negligible accuracy loss. In the proposed approach, the importance score of DNN parameters (i.e. weights) are computed using a gradient-based XAI technique called Layer-wise Relevance Propagation (LRP). Then, the scores are used to compress the DNN as follows: 1) the parameters with the negative or zero importance scores are pruned and removed from the model, 2) mixed-precision quantization is applied to quantize the weights with higher/lower score with higher/lower number of bits. The experimental results show that, the proposed compression approach reduces the model size by 64% while the accuracy is improved by 42% compared to the state-of-the-art XAI-based compression method.
Papers List
List of archived papers
Decentralized Federated Learning in IoT Environments: A Hierarchical Approach
Majid Mohammadpour - Seyedakbar Mostafavi
Designing a High Perfomance and High Profit P2P Energy Trading System Using a Consortium Blockchain Network
Poonia Taheri Makhsoos - Behnam Bahrak - Fattaneh Taghiyareh
To Transfer or Not To Transfer (TNT): Action Recognition in Still Image Using Transfer Learning
Ali Soltani Nezhad - Hojat Asgarian Dehkordi - Seyed Sajad Ashrafi - Shahriar Baradaran Shokouhi
Optimization of quantum secret sharing communication using corresponding bits
Mahsa Khorrampanah - Mohammad Bolokian - Monireh Houshmand
Real-Time Forecasting Using Mixed Frequency Time-Series Data
Armin Khayati - Mohammad Taheri - Koorush Ziarati
YOLOatt-Med: YOLO-Based Attention Mechanism for Medical Image Classification
Fatemeh Naserizadeh - Erfan Akbarnezhad Sany - Parsa Sinichi - Seyyed Abed Hosseini
A Hybrid Echo State Network for Hypercomplex Pattern Recognition, Classification, and Big Data Analysis
Mohammad Jamshidi - Fatemeh Daneshfar
Semi-automatic Detection of Persian Stopwords using FastText Library
Mohammad Dehghani - Mohammad Manthouri
Real-Time Vehicle Detection and Classification in UAV imagery Using Improved YOLOv5
Mohammad Hossein Hamzenejadi - Hadis Mohseni
Joint ADC-less Analog Demodulator and Decoder for Extended Binary (8, 4, 4) Hamming Channel Code
Mir Mahdi Safari - Jafar Pourrostam - Behzad Mozaffari Tazehkand
more
Samin Hamayesh - Version 41.5.3