DeepMaker: A multi-objective optimization framework for deep neural networks in embedded systems

Microprocessors and Microsystems - Tập 73 - Trang 102989 - 2020
Mohammad Loni1, Sima Sinaei1, Ali Zoljodi2, Masoud Daneshtalab1, Mikael Sjödin1
1School of Innovation, Design and Engineering, Mälardalen University, Västerås, Sweden
2Shiraz University of Technology, Shiraz, Iran

Tóm tắt

Từ khóa


Tài liệu tham khảo

Krizhevsky, 2012, Imagenet classification with deep convolutional neural networks, Adv. Neural Inf. Process. Syst., 1097

Hinton, 2012, Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups, IEEE Signal Process. Mag., 29, 82, 10.1109/MSP.2012.2205597

Mnih, 2015, Human-level control through deep reinforcement learning, Nature, 518, 529, 10.1038/nature14236

Zhang, 2016, Colorful image colorization, 649

T., 2015, Learning both weights and connections for efficient neural networks, Adv. Neural Inf. Process. Syst., 50, 1135

Huang, 2017, Densely connected convolutional networks, 1, 3

Yazdanbakhsh, 2017, AxBench: a multiplatform benchmark suite for approximate computing, IEEE Des. Test., 34, 60, 10.1109/MDAT.2016.2630270

Deb, 2002, A fast and elitist multiobjective genetic algorithm: NSGA-II, IEEE Trans. Evol. Comput., 6, 182, 10.1109/4235.996017

Zhang, 2016, Caffeine: towards uniformed representation and acceleration for deep convolutional neural networks, 18

F. Chollet, Keras, github, 2015. [Online]. Available:https://github.com/fchollet/keras.

Esmaeilzadeh, 2013, Power challenges may end the multicore era, Commun. ACM, 56, 93, 10.1145/2408776.2408797

Falsafi, 2017, FPGAs versus GPUs in data centers, IEEE Micro, 37, 60, 10.1109/MM.2017.19

Sharma, 2015, DNNWEAVER: from high-level deep network models to FPGA acceleration, IEEE Int. Conf. Mechatron. Electron. Autom. Eng., 76

LeCun, 1998, Gradient based learning applied to document recognition, Proc. IEEE, 86, 2278, 10.1109/5.726791

A. Krizhevsky and G. Hinton. Cifar-10 dataset. https://www.cs.toronto.edu/kriz/cifar.html.

Bergstra, 2011, Algorithms for hyperparameter optimization, 2546

Bengio, 2000, Gradient-based optimization of hyperparameters, Neural Comput., 8, 1889, 10.1162/089976600300015187

Bergstra, 2012, Random search for hyper-parameter optimization, J. Mach. Learn. Res., 281

Snoek, 2012, Practical bayesian optimization of machine learning algorithms, Adv. Neural Inf. Process. Syst., 2960

Y. Sun, B. Xue, and M. Zhang, Evolving deep convolutional neural networks for image classification (2017). arXiv:1710.10741.

B. Baker, O. Gupta, N. Naik, and R. Raskar, Designing neural network architectures using reinforcement learning (2016) arXiv Prepr 116.

B. Zoph, and Q.V. Le, Neural architecture search with reinforcement learning (2016) arXiv prepr. arXiv:1611.01578.

Z. Zhong, J. Yan, and C.L. Liu, Practical network blocks design with Q-Learning (2017) arXiv prepr. arXiv:1708.05552.

Suganuma, 2017, A genetic programming approach to designing convolutional neural network architectures, Genet. Evol. Comput. Conf., 497, 10.1145/3071178.3071229

E. Real, S. Moore, A. Selle, S. Saxena, Y.L. Suematsu, Q. Le, and A. Kurakin, Large-scale evolution of image classifiers (2017). arXiv:1703.01041.

X. Gastaldi, Shake-shake regularization (2017). arXiv:1705.07485.

He, 2016, Deep residual learning for image recognition, 770

E. Dufourq, and B.A. Bassett, EDEN: evolutionary deep networks for efficient machine learning (2017). arXiv:1709.09161.

S.H. Hasanpour, M. Rouhani, M. Fayyaz, and M. Sabokrou, Lets keep it simple, using simple architectures to outperform deeper and more complex architectures (2016). arXiv:1608.06037.

L. Wan, M. Zeiler, S. Zhang, Y. LeCun, and R. Fergus, Regularization of neural networks using dropconnect, 1 (2013) 109–111.

Simonyan, 2015, Very deep convolutional networks for large-scale image recognition, Int. Conf. Learn. Represent., 114

Grigorian, 2014, Accelerating divergent applications on simd architectures using neural networks, 317

Du, 2015, Leveraging the error resilience of neural networks for designing highly energy efficient accelerators, IEEE Trans. Comput. Des. Integr. Circt. Syst., 34, 1223, 10.1109/TCAD.2015.2419628

Moreau, 2015, SNNAP: approximate computing on programmable SOCS via neural acceleration, 603

Yazdanbakhsh, 2015, Neural acceleration for GPU throughput processors, 482

H. Li, A. K., I. Durdanovic, H. Samet, and H.P. Graf, Pruning filters for efficient convnets (2016). arXiv:1608.08710.

Han, 2015, Learning both weights and connections for efficient neural network, 1135

H. Hu, R. Peng, Y.W. Tai and C.K. Tang, Network trimming: a data-driven neuron pruning approach towards efficient deep architectures (2016) arXiv preprint arXiv:1607.03250.

Li, 2018, Filter level pruning based on similar feature extraction for convolutional neural networks, IEICE Trans. Inf. Syst., 101, 203

S. Srinivas and R.V. Babu, Data-free parameter pruning for deep neural networks (2015). arXiv:1507.06149.

Zitzler, 1998, Multiobjective optimization using evolutionary algorithms—a comparative case study, 292

Esmaeilzadeh, 2012, Dark silicon and the end of multicore scaling, IEEE Micro, 32, 10.1109/MM.2012.17

Abadi, 2016, Tensorflow: a system for large-scale machine learning, 265

Loni, 2018, ADONN: adaptive design of optimized deep neural networks for embedded systems, 397

Lu, 2019, NSGA-Net: neural architecture search using multi-objective genetic algorithm, 419

Mahdiani, 2020, ΔNN: power-efficient neural network acceleration using differential weights, IEEE Micro