I Introduction
Schizophrenia today is a chronic, frequently disabling mental disorder that affects about one per cent of the world s population [1]; And it is widely perceived as one of the most severest mental disorder compromising multiaspect of everyday quality of life [2]. This predicament often continues in spite of pharmacological treatment of psychotic symptoms [3]. Accordingly, increasing attention is paid to the studies on individual recognition in schizophrenia (IRS) with the aim of surveillance, early detection or prediagnosis.
An typical IRS scheme consists of two phases: offline training and online recognition [4, 5]. During the offline training phase, knowledgeable and unique individual characteristics are measured and recorded by some advanced brain monitoring modalities which includes EEG, electrical impedance tomography (EIT), Magnetoencephalography (MEG), Quantitative susceptibility mapping (MAP), electroneurogram (ENG), etc. These modalities promise to piece together different factors of the brain and provide new insights to help detect and treat diseases; So they are particularly appropriate for a disease as schizophrenia which impacts many aspects of the brain [6]. In this context, however, we use EEG as brain monitoring modality for the following considerations:

EEG provides a high spatiotemporal resolution data, a vivid reflection of dynamics of the brain [7].

EEG is the most inexpensive method of neuroimaging which plays a fundamental role for implementing deep learning methods [8].

The EEG of a normal and healthy brain will differ from a brain with disease or functioning abnormally or in different healthy condition [9].

EEG shows small intrapersonal differentiation and large interpersonal differentiation [10].
Note that preprocessing for these raw data is also accomplished in this phase. The ”brain fingerprinting” is then constructed for every kind of candidate. During the online recognition phase, various methods utilizing the recorded data and their extracted features can be applied to classify the candidates when the online individual characteristics were collected and refined.
Ia Related Works and Motivations
EEGbased biometrics offer an exciting new form of human computer interface where a device can be controlled and provide available data for individual recognition analysis. Related works on EEGbased individual recognition analysis are summarized briefly as follows.
So far, manuallydesigned experimental protocols and EEG features that have been commonly utilized for the devising of EEG biometric systems aimed to recognize characteristics of spatially limited sets of brain regions. Nakayama and Abe discuss the feasibility of using singlechannel EEG waveforms for singletrial classification of viewed characters [11]. Berthomier and coauthors present an automatic analysis of singlechannel frequency EEG measurements for validation in healthy individuals [12]
. The monograph gives a comprehensive study of classification of EEG signals using single channel independent component analysis, power spectrum, and linear discriminant analysis
[13]. To take advantage of spatial information provided by multichannel EEG and obtain higher classification accuracy, Krajca achieve an automatic identification of significant graphoelements in multichannel EEG recordings by adaptive segmentation and fuzzy clustering [14]. Prasad and coauthors realize a singletrial eeg classification using logistic regression based on ensemble synchronization; These works could classify each single trial of EEG as belonging to a patient with schizophrenia or a healthy control subject with 73% accuracy
[15]. However, these works are based on features extracted from single channel or multichannel EEG fragment and fail to obtain the accurate and robust classification result, thus are unfeasible for practices. Fortuately, recently advanced big data analysis based on streaming data
[16, 17, 18] could provide new means and ideas for the planning and design of classification scheme.In this paper we conduct the IRS task based on resting state EEG. There are two reasons for that. First, evidence suggests that electrical activities resting state organizes and coordinates neuronal functions
[19]. Second, certain tasks cannot be performed by certain group of people, e.g., schizophrenia, Attention Deficit Disorder, or hyperactivity disorder [20].The difficulty encountered in resting state EEG based IRS scheme is that resting state EEG [21]
streams lack taskrelated feature, thus leading to a hard task to obtain the best and unique feature for an individual. Accordingly, there has been emerging a great need for the capability to extract features automatically. Kottaimalai and coauthors put forward EEG signal classification using Principal Component Analysis with Neural Network in Brain Computer Interface applications
[22]. Li and Fan suggest a classification method to separate Schizophrenia and depression by EEG with artificial neural networks (ANN) [23]. Ruffini et al. present EEGdriven classification for Prognosis of Neurodegeneration in AtRisk Patients by recurrent neural networks (RNN)
[24]. ANNbased and RNNbased neural network structures require the nonvectorial inputs such as matrices to be converted into vectors which has been proved of problematic
[25, 26]. The vectorization of EEG streams would lose spatiotemporal information and give a very large solution space that demands very special treatments to the network parameters and high computational cost. As novel alternatives, convolution neural network (CNN) can help improve a learning system with three advantages sparse interactions, parameter sharing and equivariant representations
[27]. Recently, Ma et al. conduct resting state EEGbased biometrics for individual identification using convolutional neural networks [28]; And their results indicate that the CNNbased jointoptimized EEGbased biometric system yields a high degree of accuracy of identification (88%) which still can not reach the practical requirement. In summary, to obtain a higher classification accuracy, combining CNNbased network structure with spatiotemporal EEG analysis will be indispensable.IB Our Contributions
Based on considerations above, we propose a new IRS scheme using advanced deep learning methods, aiming at automatically extracting features and performing classification. Our main contributions are summarized as follows.

Instead of utilizing short term EEG data which were found insufficient to provide required information for IRS analysis, we employ streaming EEG data collected by multichannel scalp electrodes.

Three kinds of advanced deep learning methods were developed for IRS analysis.

The classifier which was widely used in classical deep learning methods is replaced by RF with aim of improving classification accuracy.

To tackle the classification problem with EEG data streams, a voting layer is developed at the top of the employed neural networks.

Various experiments are conducted to investigate the effectiveness and robustness of proposed IRS scheme.
The remainder of this paper is structured as follows. Section II firstly introduces the procedure of collection, preprocessing and mathematical representation for EED data streams; The proposed IRS scheme based on advanced deep learning methods is then developed in the second part of Section II. In Section III, numerical case studies are provided to evaluate the performance of the proposed IRS scheme. Conclusion and acknowledgement of this research is given in Section IV and Section V, respectively.
Ii Materials and Methods
Iia EEG Data Collection
The present work aimed to study IRS issue by assessing three types of subjects: characteristics in high risk (CHR) individuals, clinically stable first episode patients with schizophrenia (FES) and healthy controls (HC). 120 subjects (40 CHRs, 40 FESs and 40 HCs) were included; And all subjects to be investigated in this context were recruited from outpatients at Shanghai Mental Health Center. All subjects were free of mental retardation, neurological diseases, substance abuse or alcohol and any physical illness that may influence their cognitive function. The study protocol was approved by the Institutional Review Board of Shanghai Mental Health Center, and informed consents were obtained from all the subjects.
The experimental data were provided by Department of EEG Source Imaging in Shanghai Mental Health Center. So the data collection process is same with their previous work [29]. Participants were seated 1 m from the screen in a sound attenuated and electrically shielded chamber with dim illumination. EEGs were recorded from 64channel scalp electrodes mounted in an elastic cap (BrainCap, Brain Products, Inc., Bavaria, Germany) including two pairs of vertical and horizontal electrooculography (EOG) electrodes. The electrodescalp impedance was kept below 5 k for each electrode. Our analysis was performed on eyeopen resting conditions, each single recording lasting over 300 seconds in time. Data recording was referenced to the tip of nose and sampled at Hz.
IiB EEG Data Preprocessing and Mathematical Representation
The brain vision analyzer (1.05, Brain Products, Inc.) was utilized for EEG preprocessing [30]. Artifacts caused by vertical and horizontal eyes movements and blinks were removed offline by an ocular correction algorithm [31]. All the artifactreduced EEG data were referenced using the common average reference, bandpass filtered into 0.01 C50 Hz using a zero phaseshift IIR filter (24 dB/Oct). See Fig. 1 and Fig. 2 for an illustration. After that, the broadband EEG signals containing artifacts were excluded using EEGLAB [31]. After the preprocessing, in addition to preserve the complete signals, the EEG signals were also bandpass filtered into four classic frequency bands, i.e., Hz, Hz, Hz, Hz and Hz bands, respectively, using leastsquares FIR filters [32].
In order to facilitate subsequent analysis, the mathematical representation of filtered EEG streams are described in the following. Let and denote the number of the available channel number of scalp electrodes and sampling time, respectively. To ensure the same length of collected EEG data in the following analysis, we have for all subjects. The total length of EEG data collected at every scalp electrode is . For th type subject, a sliding window based data allocation scheme for the large EEG data matrix is presented as follows. Let be the sliding window size and , then a sequence of matrix
(1) 
is obtained to represent the collected EEG data streams. As shown in Fig. 3, these fragments are considered as raw Brain fingerprinting of all subjects.
IiC Proposed IRS Scheme based on Advanced Deep Learning Methods
IiC1 Classical Deep Learning Structures
As introduced in the Section IA, we have introduced previous biometrics that embrace classical deep learning methods, such as ANN, RNN, CNN and their modified versions. Here, some technical details are discussed in order to get better understanding on how to apply them. Input, a kind of network and classifier contribute to the basic elements of a deep learning structure (See Fig. 4). The implementation of classical deep learning methods for IRS problem is discussed in the following.
As introduced in the Section IIB, the collected EEG streams recorded for offline analysis are represented by data fragments , where , , . These fragments were utilized to train the neural networks (ANN, CNN and RNN) after local normalization scheme represented by
(2) 
For IRS classification problems based on deep learning methods, it is standard to use classifier at the top. Let the subjects be a finite space with a finite observation space . Let
be the learned model of the conditional probability of seeing observation data
with people . Let and be the activation of the penultimate layer nodes and the weight connecting the penultimate layer to a classifier layer, respectively. The total input into the classifier layer, denoted by , is(3) 
Given the classifier, we have
(4) 
The predicted class for the single fragment would be
(5) 
IiC2 Deep Learning Methods using RF Classifier
Most deep learning methods for classification utilizing convolutional and fullconnected layers have used classifier to learn the small size parameters. There are exceptions, significantly in works [33, 34], supervised embedding with nonlinear NCA [35], semisupervised deep embedding [36]
and deep learning using linear support vector machines
[37]. In this paper, we replace the with RF for classification. RF has been studied extensively in the fields of nonparametric statistics [38] and continue to be very popular because of its simplicity and because it is very successful for many practical problems [39, 40]. Here we firstly summarize the basic principle of RF as follows.Let and be the data features and the corresponding labels. RF is built from a training set that make predictions for new points by looking at the neighborhood of the point, formalized by a weight function in th tree:
(6) 
Here, is the nonnegative weight of the th training point relative to the new point and is the number of nodes in the penultimate layer of the neural network in this work. For any , the weights sum to one. Since a forest averages the predictions of a group of trees with individual weight functions , its predictions are
(7) 
then the predicted class is
(8) 
For more technical details about RF, interested readers are referred to the distinguished works by Breiman [41, 42].
Another advanced classifier, a linear multiclass support vector machine (mSVM) which has been proved of higher classification accuracy in [43, 44], is also adopted for the purposes of comparison. We verified the effectiveness of the proposed scheme on the wellknown dataset: a ninelayer CNN achieved a test error with RF classifier, with mSVM classifier and with classifier.
IiC3 Streaming EEG Classification with a Voting Scheme
It is worth noting that the above modified deep learning methods are suitable to classify subjects with single fragment . Let . To handle the scenario that the subjects are with EEG streams , we develop a voting layer whose decision rule is denoted by
(9) 
where .
Moreover, in this paper, we also adopt the some other recently developed techniques to improve the performance of deep learning methods employed in the IRS analysis. Specially, we use exponential linear unit (ELU) proposed in [45] to accelerate the learning speed in deep neural networks. The technique [46] is utilized to prevent substantial overfitting problem.
Lastly, for the readers’ convenience, we give a brief summary of the modified deep learning methods employed in this context in the following two tables (Tab. I and Tab. II).
CNN  ANN  RNN  

1st  Input  Input  Input 
2nd  2(Conv. + ELU) (kernel: , stride:2) 
Vectorization  Vectorization 
3rd  Max Pooling (kernel:); Dropout(Rate:0.25)  Dense(512)  recurrent layer (hidden units = 100) 
4th  2(Conv. + ELU) (kernel:)  Activation(’relu’) 
Dense(3) 
5th  Max Pooling (kernel:, stride:2); Dropout(Rate:0.25)  Dropout(Rate:0.25)  or mSVM or RF 
6th  2(Conv. + ELU) (kernel:)  Dense(512)  Output(Predict) 
7th  Max Pooling (kernel:, stride:2); Dropout(Rate:0.25)  Activation(’relu’)   
8th  Dense(128) + ELU  Dropout(Rate:0.25)   
9th  Dropout(Rate:0.5)  Dense(512)   
10th  Dense(128) + ELU  Activation(’relu’)   
11th  Dropout(Rate:0.5)  Dropout(Rate:0.25)   
12th  Dense(3) + ELU  Dense(3)   
13th  Dropout(Rate:0.5)  Dropout(Rate:0.5)   
14th  or mSVM or RF  or mSVM or RF   
15th  Output(Predict)  Output(Predict)   
Deep Learning Methods  Explanation 

ANNV  Classical ANN with classifier and a voting layer 
RNNV  Classical RNN with classifier and a voting layer 
CNNV  Classical CNN with classifier and a voting layer 
ANNV+mSVM  Modified ANN using mSVM classifier and a voting layer 
RNN+mSVM  Modified ANN using mSVM classifier and a voting layer 
CNN+mSVM  Modified ANN using mSVM classifier and a voting layer 
ANN+RF  Modified ANN utilizing RF classifier and a voting layer 
RNN+RF  Modified RNN utilizing RF classifier and a voting layer 
CNN+RF  Modified CNN utilizing RF classifier and a voting layer 
Iii Case Studies and Discussion
In this section, various experiments are developed to evaluate the performance of the proposed IRS schemes. We use the cross validation method to evaluate the performance of the proposed IRS scheme. Our results are the averages of 1000 independent run on GeForce GTX 750.
Iiia The Accuracy of Timedomain and Frequencydomain EEG Data Streams
Timedomain (as introduced in Section. IIB
) and Frequencydomain (Amplitude of Fourier transform) EEG Data Streams are utilized firstly to perform and report accuracy assessments of proposed IRS schemes. The window size are set as
. The test data size are kept same with the training data for every subject. Tab. III shows that the proposed CNNV+RF has the best classification accuracy against other methods.FES  HC  CHR  FES  HC  CHR  

ANNV  0.809  0.831  0.622  0.807  0.824  0.631 
RNNV  0.742  0.803  0.594  0.731  0.792  0.588 
CNNV  0.923  0.952  0.755  0.915  0.949  0.749 
ANNV+mSVM  0.811  0.841  0.643  0.804  0.929  0.639 
RNNV+mSVM  0.759  0.826  0.602  0.744  0.807  0.589 
CNNV+mSVM  0.946  0.983  0.790  0.937  0.985  0.766 
ANNV+RF  0.827  0.846  0.657  0.813  0.839  0.655 
RNNV+RF  0.766  0.839  0.613  0.793  0.816  0.649 
CNNV+RF  0.967  0.992  0.816  0.955  0.981  0.799 
Iv Conclusion
In conclusion, we have shown that CNNVRF performs better than and CNNVmSVM on a wellknown dataset () and resting state EEG streams used in this paper. Switching from or mSVM to RF is incredibly simple and appears ro be helpful for classification problems. The experimental results show that the classification performance would be improved as the size of training and data database becomes larger. In the future, the proposed biometrics system should be tested on a larger group and more classes of subjects, providing further identification of accuracy, robustness and applicability of the system. The experiments also suggest that our results can be improved simply by waiting for faster GPUs.
V Acknowledgements
We are appreciated for department of EEG source imaging leaded by professor Jijun Wang and professor Chunbo Li from SHJC for data providing and discussion. We also grateful to Dr. Tianhong Zhang from SHJC for his expert collaboration on data analysis.
References
 [1] T. R. Insel, “Rethinking schizophrenia.” Nature, vol. 468, no. 7321, pp. 187–193, 2010.
 [2] S. Aswin, A. R. Bialas, D. R. Heather, D. Avery, T. R. Hammond, K. Nolan, T. Katherine, P. Jessy, B. Matthew, and V. D. Vanessa, “Schizophrenia risk from complex variation of complement component 4:,” Nature, vol. 530, no. 7589, p. 177, 2016.
 [3] R. A. Gould, K. T. Mueser, E. Bolton, V. Mays, and D. Goff, “Cognitive therapy for psychosis in schizophrenia: an effect size analysis.” Schizophrenia Research, vol. 48, no. 2 C3, pp. 335–342, 2001.
 [4] A. Ulas, U. Castellani, V. Murino, M. Bellani, M. Tansella, and P. Brambilla, Heat Diffusion Based Dissimilarity Analysis for Schizophrenia Classification. Springer Berlin Heidelberg, 2011.
 [5] H. S. Choi, B. Lee, and S. Yoon, “Biometric authentication using noisy electrocardiograms acquired by mobile sensors,” IEEE Access, vol. 4, pp. 1266–1273, 2016.
 [6] A. E. Hassanien and A. T. Azar, Brain Computer Interfaces: Current Trends and Applications, 2014.
 [7] C. Guger, A. Schlogl, C. Neuper, D. Walterspacher, T. Strein, and G. Pfurtscheller, “Rapid prototyping of an eegbased bci,” IEEE Trans.on Neural Systems and Rehab.eng, 2001.
 [8] C. P. Panayiotopoulos, EEG and brain imaging. Springer London, 2010.
 [9] M. G. Knyazeva and G. M. Innocenti, “Eeg coherence studies in the normal brain and after earlyonset cortical pathologies,” Brain Research Reviews, vol. 36, no. 2 C3, pp. 119–128, 2001.
 [10] M. A. Guevara, I. Lorenzo, C. Arce, J. Ramos, and M. CorsiCabrera, “Inter and intrahemispheric eeg correlation during sleep and wakefulness.” Sleep, vol. 18, no. 4, p. 257, 1995.
 [11] M. Nakayama and H. Abe, “Feasibility of using singlechannel eeg waveforms for singletrial classification of viewed characters,” in International Conference on Information Society, 2010, pp. 218–223.
 [12] C. Berthomier, X. Drouot, M. HermanStoica, P. Berthomier, J. Prado, D. BokarThire, O. Benoit, J. Mattout, and M. P. D’Ortho, “Automatic analysis of singlechannel sleep eeg: validation in healthy individuals.” Sleep, vol. 30, no. 11, pp. 1587–1595, 2007.
 [13] H. Tjandrasa and S. Djanali, Classification of EEG Signals Using Single Channel Independent Component Analysis, Power Spectrum, and Linear Discriminant Analysis, 2016.
 [14] V. Krajca, S. Petranek, I. Patakova, and A. Varri, “Automatic identification of significant graphoelements in multichannel eeg recordings by adaptive segmentation and fuzzy clustering.” International Journal of BioMedical Computing, vol. 28, no. 28, pp. 71–89, 1991.
 [15] P. D. Prasad, H. N. Halahalli, J. P. John, and K. K. Majumdar, “Singletrial eeg classification using logistic regression based on ensemble synchronization.” IEEE Journal of Biomedical and Health Informatics, vol. 18, no. 3, pp. 1074–1080, 2014.
 [16] E. Landhuis, “Neuroscience: Big brain, big data,” Nature, 2017.
 [17] L. Chu, R. Qiu, X. He, Z. Ling, and Y. Liu, “Massive streaming pmu data modeling and analytics in smart grid state evaluation based on multiple highdimensional covariance tests,” 2017, accepted by IEEE TRANSACTIONS ON BIG DATA.
 [18] R. C. Qiu and P. Antonik, Smart Grid and Big Data: Theory and Practice. Hoboken, NJ, USA: John Wiley and Sons, 2015.
 [19] S. Kuhn and J. Gallinat, “Restingstate brain activity in schizophrenia and major depression: a quantitative metaanalysis.” Schizophrenia Bulletin, vol. 39, no. 2, pp. 358–65, 2011.
 [20] C. Karatekin and R. F. Asarnow, “Working memory in childhoodonset schizophrenia and attentiondeficit/hyperactivity disorder.” Psychiatry Research, vol. 80, no. 2, pp. 165–176, 1998.
 [21] S. R. Sponheim, B. A. Clementz, W. G. Iacono, and M. Beiser, “Resting eeg in firstepisode and chronic schizophrenia.” Psychophysiology, vol. 31, no. 1, pp. 37–43, 1994.
 [22] R. Kottaimalai, M. P. Rajasekaran, V. Selvam, and B. Kannapiran, EEG signal classification using Principal Component Analysis with Neural Network in Brain Computer Interface applications, 2013.
 [23] Y. J. Li and F. Y. Fan, “Classification of schizophrenia and depression by eeg with anns,” in International Conference of the Engineering in Medicine and Biology Society, 2006, pp. 2679–2682.
 [24] G. Ruffini, D. Ibanez, M. Castellano, S. Dunne, and A. SoriaFrisch, EEGdriven RNN Classification for Prognosis of Neurodegeneration in AtRisk Patients. Springer International Publishing, 2016.
 [25] J. Gao, Y. Guo, and Z. Wang, “Matrix neural networks,” CoRR, vol. abs/1601.03805, 2016. [Online]. Available: http://arxiv.org/abs/1601.03805
 [26] J. Schmidhuber, “Deep learning in neural networks: An overview.” Neural Networks, vol. 61, pp. 85–117, 2015.
 [27] Y. Lecun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015.
 [28] L. Ma, J. W. Minett, T. Blu, and S. Y. Wang, “Resting state eegbased biometrics for individual identification using convolutional neural networks,” in International Conference of the IEEE Engineering in Medicine and Biology Society, 2015, p. 2848.
 [29] J. Sun, Y. Tang, K. O. Lim, J. Wang, S. Tong, H. Li, and B. He, “Abnormal dynamics of eeg oscillations in schizophrenia patients on multiple time scales.” IEEE transactions on biomedical engineering, vol. 61, no. 6, pp. 1756–1764, 2014.
 [30] M. M. Lorist, E. Bezdan, M. T. Caat, M. M. Span, J. B. T. M. Roerdink, and N. M. Maurits, “The influence of mental fatigue and motivation on neural network dynamics; an eeg coherence study,” Brain Research, vol. 1270, no. 2, pp. 95–106, 2009.
 [31] C. Brunner, A. Delorme, and S. Makeig, “Eeglab  an open source matlab toolbox for electrophysiological research.” Biomedizinische Technik and biomedical Engineering, vol. 58, no. 18, pp. 3234–3234, 2013.
 [32] P. P. Vaidyanathan and T. Nguyen, “Nguyen, t.q.: Eigenfilters: A new approach to least squares fir filter design and applications including nyquist filters. ieee trans. circuits syst. 34, 1123,” IEEE Transactions on Circuits and Systems, vol. 34, no. 1, pp. 11–23, 1987.
 [33] J. Weston, F. Ratle, H. Mobahi, and R. Collobert, Deep Learning via Semisupervised Embedding. Springer Berlin Heidelberg, 2012.
 [34] J. Weston, F. Ratle, and R. Collobert, “Lecture notes in computer science,” in International Conference, 2008, pp. 1168–1175.
 [35] G. W. Taylor, R. Fergus, G. Williams, I. Spiro, and C. Bregler, “Posesensitive embedding by nonlinear nca regression.” in Advances in Neural Information Processing Systems 23: Conference on Neural Information Processing Systems 2010. Proceedings of A Meeting Held 69 December 2010, Vancouver, British Columbia, Canada, 2010, pp. 2280–2288.

[36]
J. Weston, F. Ratle, and R. Collobert, “Deep learning via semisupervised
embedding,” in
International Conference on Machine Learning
, 2008, pp. 1168–1175.  [37] C. Gentile, “A new approximate maximal margin classification algorithm,” Journal of Machine Learning Research, vol. 2, no. 2, pp. 213–242, 2001.

[38]
R. Genuer, “Variance reduction in purely random forests,”
Journal of Nonparametric Statistics, vol. 24, no. 3, pp. 1–20, 2012.  [39] M. Pal, “Random forest classifier for remote sensing classification,” International Journal of Remote Sensing, vol. 26, no. 1, pp. 217–222, 2005.
 [40] R. D azuriarte and d. A. S. Alvarez, “Gene selection and classification of microarray data using random forest.” Bmc Bioinformatics, vol. 7, no. 1, p. 3, 2006.
 [41] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001.
 [42] ——, “Consistency for a simple model of random forests,” Applied Economics  APPL ECON, 2004.
 [43] R. Arunkumar and P. Karthigaikumar, “Multiretinal disease classification by reduced deep learning features,” Neural Computing and Applications, pp. 1–6, 2015.
 [44] Y. Tang, “Deep learning using linear support vector machines,” Computer Science, 2015.

[45]
A. Shah, E. Kadam, H. Shah, S. Shinde, and S. Shingade, “Deep residual
networks with exponential linear unit,” in
International Symposium on Computer Vision and the Internet
, 2016, pp. 59–65. 
[46]
A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” in
International Conference on Neural Information Processing Systems, 2012, pp. 1097–1105.
Comments
There are no comments yet.