Training Deep Neural Networks with Constrained Learning Parameters

被引:0
|
作者
Date, Prasanna [1 ]
Carothers, Christopher D. [1 ]
Mitchell, John E. [2 ]
Hendler, James A. [1 ]
Magdon-Ismail, Malik [1 ]
机构
[1] Rensselaer Polytech Inst, Dept Comp Sci, Troy, NY 12180 USA
[2] Rensselaer Polytech Inst, Dept Math Sci, Troy, NY 12180 USA
关键词
Deep Neural Networks; Training Algorithm; Deep Learning; Machine Learning; Artificial Intelligence; LOCAL SEARCH; OPTIMIZATION;
D O I
10.1109/ICRC2020.2020.00018
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Today's deep learning models are primarily trained on CPUs and GPUs. Although these models tend to have low error, they consume high power and utilize large amount of memory owing to double precision floating point learning parameters. Beyond the Moore's law, a significant portion of deep learning tasks would run on edge computing systems, which will form an indispensable part of the entire computation fabric. Subsequently, training deep learning models for such systems will have to be tailored and adopted to generate models that have the following desirable characteristics: low error, low memory, and low power. We believe that deep neural networks (DNNs), where learning parameters are constrained to have a set of finite discrete values, running on neuromorphic computing systems would be instrumental for intelligent edge computing systems having these desirable characteristics. To this extent, we propose the Combinatorial Neural Network Training Algorithm (CoNNTrA), that leverages a coordinate gradient descent-based approach for training deep learning models with finite discrete learning parameters. Next, we elaborate on the theoretical underpinnings and evaluate the computational complexity of CoNNTrA. As a proof of concept, we use CoNNTrA to train deep learning models with ternary learning parameters on the MNIST, Iris and ImageNet data sets and compare their performance to the same models trained using Backpropagation. We use following performance metrics for the comparison: (i) Training error; (ii) Validation error; (iii) Memory usage; and (iv) Training time. Our results indicate that CoNNTrA models use 32x less memory and have errors at par with the Backpropagation models.
引用
收藏
页码:107 / 115
页数:9
相关论文
共 50 条
  • [1] Entropy-Constrained Training of Deep Neural Networks
    Wiedemann, Simon
    Marban, Arturo
    Mueller, Klaus-Robert
    Samek, Wojciech
    2019 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2019,
  • [2] Constrained Training of Recurrent Neural Networks for Automata Learning
    Aichernig, Bernhard K.
    Koenig, Sandra
    Mateis, Cristinel
    Pferscher, Andrea
    Schmidt, Dominik
    Tappler, Martin
    SOFTWARE ENGINEERING AND FORMAL METHODS, SEFM 2022, 2022, 13550 : 155 - 172
  • [3] TRAINING STABLE GRAPH NEURAL NETWORKS THROUGH CONSTRAINED LEARNING
    Cervino, Juan
    Ruiz, Luana
    Ribeiro, Alejandro
    2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2022, : 4223 - 4227
  • [4] Dirichlet Energy Constrained Learning for Deep Graph Neural Networks
    Zhou, Kaixiong
    Huang, Xiao
    Zha, Daochen
    Chen, Rui
    Li, Li
    Choi, Soo-Hyun
    Hu, Xia
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34
  • [5] Learning to Optimize with Unsupervised Learning: Training Deep Neural Networks for URLLC
    Sun, Chengjian
    Yang, Chenyang
    2019 IEEE 30TH ANNUAL INTERNATIONAL SYMPOSIUM ON PERSONAL, INDOOR AND MOBILE RADIO COMMUNICATIONS (PIMRC), 2019, : 451 - 457
  • [6] An active learning framework for adversarial training of deep neural networks
    Susmita Ghosh
    Abhiroop Chatterjee
    Lance Fiondella
    Neural Computing and Applications, 2025, 37 (9) : 6849 - 6876
  • [7] Learning to Optimize: Training Deep Neural Networks for Interference Management
    Sun, Haoran
    Chen, Xiangyi
    Shi, Qingjiang
    Hong, Mingyi
    Fu, Xiao
    Sidiropoulos, Nicholas D.
    IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2018, 66 (20) : 5438 - 5453
  • [8] Learning Not to Learn: Training Deep Neural Networks with Biased Data
    Kim, Byungju
    Kim, Hyunwoo
    Kim, Kyungsu
    Kim, Sungjin
    Kim, Junmo
    2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019), 2019, : 9004 - 9012
  • [9] Adaptive Learning Rate and Momentum for Training Deep Neural Networks
    Hao, Zhiyong
    Jiang, Yixuan
    Yu, Huihua
    Chiang, Hsiao-Dong
    MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2021: RESEARCH TRACK, PT III, 2021, 12977 : 381 - 396
  • [10] The Evolution of Training Parameters for Spiking Neural Networks with Hebbian Learning
    Kozdon, Katarzyna
    Bentley, Peter
    2018 CONFERENCE ON ARTIFICIAL LIFE (ALIFE 2018), 2018, : 276 - 283