=Paper=
{{Paper
|id=Vol-2771/paper30
|storemode=property
|title= Neural Architecture Search using Particle
Swarm and Ant Colony Optimization
|pdfUrl=https://ceur-ws.org/Vol-2771/AICS2020_paper_30.pdf
|volume=Vol-2771
|authors=Séamus Lankford,Diarmuid Grimes
|dblpUrl=https://dblp.org/rec/conf/aics/LankfordG20
}}
== Neural Architecture Search using Particle
Swarm and Ant Colony Optimization==
https://ceur-ws.org/Vol-2771/AICS2020_paper_30.pdf
Neural Architecture Search using Particle
Swarm and Ant Colony Optimization
Séamus Lankford1 and Diarmuid Grimes2
1
Adapt Centre, Dublin City University, Ireland
seamus.lankford@adaptcentre.ie
2
Cork Institute of Technology, Ireland
diarmuid.grimes@cit.ie
Abstract. Neural network models have a number of hyperparameters
that must be chosen along with their architecture. This can be a heavy
burden on a novice user, choosing which architecture and what values
to assign to parameters. In most cases, default hyperparameters and ar-
chitectures are used. Significant improvements to model accuracy can
be achieved through the evaluation of multiple architectures. A process
known as Neural Architecture Search (NAS) may be applied to automat-
ically evaluate a large number of such architectures.
A system integrating open source tools for Neural Architecture Search
(OpenNAS), in the classification of images, has been developed as part
of this research. OpenNAS takes any dataset of grayscale, or RBG im-
ages, and generates Convolutional Neural Network (CNN) architectures
based on a range of metaheuristics using either an AutoKeras, a transfer
learning or a Swarm Intelligence (SI) approach.
Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO)
are used as the SI algorithms. Furthermore, models developed through
such metaheuristics may be combined using stacking ensembles.
In the context of this paper, we focus on training and optimizing CNNs
using the Swarm Intelligence (SI) components of OpenNAS. Two major
types of SI algorithms, namely PSO and ACO, are compared to see which
is more effective in generating higher model accuracies. It is shown, with
our experimental design, that the PSO algorithm performs better than
ACO. The performance improvement of PSO is most notable with a more
complex dataset. As a baseline, the performance of fine-tuned pre-trained
models is also evaluated.
Keywords: AutoML · NAS · Swarm Intelligence · PSO · ACO · CNN
1 Introduction
The area of Auto Machine Learning (AutoML) [1] is a growing area of interest in
recent years. This is reflected in the development of several open source AutoML
libraries among which include Auto-WEKA [2], Hyperopt-Sklearn [3], AutoKeras
[4], Auto-Sklearn [5, 6] and TPOT [7].
Despite a renewal of interest in AutoML, many of these open source solutions
focus on creating simpler neural architectures. Libraries which concentrate on
Copyright 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
2 S. Lankford and D. Grimes
generating more complex architectures, such as CNNs, are at early stages of
development. Consequently they are poorly documented and often unreliable
[4]. In addition, the alternative of using commercial platforms is expensive and
therefore users are left with few practical or viable options.
The development of OpenNAS integrates several metaheuristic approaches
in a single application used for the neural architecture search of more complex
neural architectures such as convolutional neural networks. Furthermore, the
effectiveness of NAS in generating good neural architectures for image classifica-
tion is evaluated. Standard approaches to NAS, using the AutoKeras framework,
are also incorporated into the system design.
A key aspect of the study is to contrast Swarm Intelligence (SI) algorithms
for NAS. Consequently, Particle Swarm Optimization (PSO) [8] and Ant Colony
Optimization (ACO) [9] have been chosen as metaheuristics for creating high
performing CNN architectures for grayscale and RGB image datasets.
2 Background
2.1 Convolutional Neural Networks
CNNs are feed-forward Deep Neural Networks (DNNs) used for image recogni-
tion. The original CNN architecture was proposed by LeCun [10] and consisted
of two convolution layers, two pooling layers, two fully connected (FC) layers
and an output layer. Subsequently, numerous models were developed including
popular ones such as ResNet [11] and VGG [12]. In this study, custom CNN
architectures are created by using SI heuristics to find better combinations of
convolutional, pooling and FC layers.
2.2 Auto ML
AutoML involves the automation of the entire machine learning pipeline in-
cluding data augmentation, feature engineering, model selection, choice of hyper
parameters and finally neural architecture selection and creation. By constrast,
NAS has a more narrow focus in that it concentrates on neural architecture
selection and creation [13].
Tree-based Pipeline Optimization Tool (TPOT) is an open source python
package that uses genetic programming in optimizing the machine learning
pipeline [7]. The library performs well on simple NAS tasks involving the scikit-
learn API. Given this study involves generating more complex CNNs, rather
than developing optimal pipelines, it was decided not to use TPOT as part of
the initial solution architecture. However, as part of future work, it may have a
role in optimizing hyper parameter selection.
AutoKeras [4] is an open source AutoML system using Bayesian optimization
and network morphism for efficient neural architecture search.
OpenNAS using SI 3
2.3 Neural Architecture Search
Neural architecture search is the process of automatically finding and tuning
DNNs. It has been shown that DNNs have made remarkable progress in solving
many real world problems such as image recognition, speech recognition and
machine translation [14]. In general, NAS systems consist of three main com-
ponents: a search space, a search algorithm and an evaluation strategy. The
search space sets out which architectures can be used in principle whereas the
search strategy outlines how the search space is explored. Finally the evaluation
strategy determines which architectures yield the best results on unseen data.
A basic approach to NAS is the brute force training and evaluation of all
possible model combinations. On completion, the best performing model is se-
lected. However, this is impractical due to the combinatorics of the problem.
Using metaheuristics, such as swarm intelligence, is an alternative which seeks
the best model within reasonable time constraints.
2.4 Swarm Intelligence
Swarm Intelligence, a category of Evolutionary Computing, has been used for
classification problems in the following forms: Particle Swarm Optimization
(PSO) [15, 16] and Ant Colony Optimization (ACO) [17].
Particle Swarm Optimization PSO belongs to the class of swarm intelli-
gence techniques and is a population-based stochastic technique for solving op-
timization problems developed in 1995. An open source python library, for CNN
optimization using the PSO algorithm was developed by Fernandes et al [18].
The results demonstrate that their approach, psoCNN, quickly finds CNN ar-
chitectures which offer competitive performance for any given dataset.
Ant Colony Optimization ACO, modelled on the activities of real ant colonies,
involves moving through a parameter space of all potential solutions to find the
optimal weights for a neural network.
Using ACO, a system known as DeepSwarm was developed by Byla and Pang
[19] to find high performing neural architectures for CNNs. They showed that it
offers competitive performance when tested on well-known image datasets.
3 Approach
With artificial neural networks, there are many parameters to choose from such
as the number of hidden network layers, number of neurons per layer, type of
activation function, choice of optimizer and so on. The final network design often
depends on the problem domain and is typically achieved in a time consuming
trial and error fashion.
Similar problems exist with CNNs but these problems are exacerbated by
the length of time, and amount of computational resources required to train
4 S. Lankford and D. Grimes
such networks. Clearly, a core objective of NAS is to find good network perfor-
mance within acceptable time limits through the reduction of both the number
of networks tested and the length of time required for their evaluation. The im-
plementation of NAS can be achieved through a variety of approaches including
transfer learning using pre-trained networks, network morphism or swarm in-
telligence. Using these approaches as its pillars, a NAS system (OpenNAS) has
been built which tackles such problems 3 . OpenNAS does not enforce a partic-
ular architecture but rather it allows novel and interesting architectures to be
discovered.
In this work we focus on the swarm intelligence component of the OpenNAS
system. The swarm optimization techniques currently used are Particle Swarm
Optimization and Ant Colony Optimization.
Table 1. Parameters for Particle Swarm Optimization
Config A Config B
Swarm
Number of iterations 10 20
Swarm size 20 10
Cg 0.5 0.5
CNN architecture
Minimum outputs from a Conv layer 3 3
Maximum outputs from a Conv layer 256 256
Maximum neurons in a FC layer 300 300
Minimum size of a Conv kernel 3x3 3x3
Maximum size of a Conv kernel 7x7 7x7
Minimum layers 3 3
Maximum layers 20 20
CNN Training
# epochs for particle evaluation 5 5
# epochs for global best 100 100
Dropout rate 0.5 0.5
Batch normalize layer outputs Yes Yes
Probability Settings
probability convolution 0.6 0.6
probability pooling 0.3 0.3
probability fully connected 0.1 0.1
The PSO algorithm determines how the principal CNN layer types, and their
associated hyperparameters, are connected together. The generated models con-
sist of architectures using a mix of convolutional, average pooling, max pooling
and fully connected layers. In addition, dropout layers and batch normalization
layers are also added to alleviate overfitting. The hyperparameters associated
with each layer type are indicated in Table 1.
3
https://github.com/seamusl/OpenNAS-v1
OpenNAS using SI 5
Particle architectures, i.e. model architectures, are compiled for a number of
epochs and evaluation is carried out using the standard loss function of cross-
entropy loss. Particle architectures with the smallest loss are selected by the
algorithm. The number of epochs parameter for pBest must be carefully chosen
since it is the main driver of both run time and model accuracy.
Using an ACO approach, the parameters used for model training in the explo-
ration process are highlighted in Table 2. Two test configurations are considered.
In the first case, 8 ants are used with 30 epochs and in the second case, 16 ants
are used with 15 epochs. The depth parameter was fixed at 20.
Table 2. Parameters for Ant Colony Optimization
Config A Config B
Ant Colony
Number of Ants 8 16
Number of Epochs 30 15
Search Depth 20 20
CNN architecture
Kernel Sizes 1, 3, 5 1, 3, 5
Minimum layers 1 1
Maximum layers 20 20
CNN Training
Dropout rate 0.1, 0.3, 0.5 0.1, 0.3, 0.5
Batch normalize layer outputs Yes Yes
Probability Settings
pheromone start, decay, evaporation 0.1 0.1
greediness 0.5 0.5
Fine tuning was implemented by initially removing the fully connected layers
from the top of the model. Two blocks are then added, each of which has a fully
connected layer, a batch normalization layer and a dropout layer. The hybrid
structure is then trained with the new dataset. Fine tuning of a VGG16 network
is illustrated in Figure 1.
Fig. 1. Tuned VGG16 model
6 S. Lankford and D. Grimes
Fig. 2. Open source Neural Architecture Search (OpenNAS) System Design
4 Design
The high level view of the system architecture is presented in Figure 2. The
system is organized into the following python modules: OpenNAS, pre-processor,
trainer, ensemble, super stacker, sysconfig and loader. The pre-train function
uses transfer learning as either a feature extractor or to fine tune the pre-trained
networks of VGG16, VGG19, MobileNet or ResNet50.
With the swarm function, PSO or ACO can be used to search for the best
neural architecture. Existing open source python libraries were customized for
both PSO and ACO functionality. Particle swarms were implemented using a
psoCNN library [18] whereas ant colonies used the DeepSwarm library [19]. The
environment required Python 3.7, Tensorflow 1.14, Keras 2.2.4, Numpy 1.16.4
and Matplotplib 3.1.0.
Existing NAS tools, such as AutoKeras, were also integrated into the Open-
NAS system. AutoKeras is a powerful open source library which provides func-
tions to automatically search for optimal architectures for deep learning models.
OpenNAS using SI 7
However, this library is still in beta development and the associated documen-
tation is quite poor.
With the ensemble module, there are options to build stacked ensembles using
either homogeneous or heterogeneous base learners. These learner outputs are
subsequently passed to a suite of meta learner algorithms. The system generates
the optimal neural architecture model using the chosen heuristic.
5 Evaluation
Two datasets were chosen for the experimental design, namely CIFAR10 [20]
and Fashion Mnist [21]. A primary research objective is the development of a
Neural Architecture Search tool which generates high performing architectures
for generic datasets of either grayscale (one channel) or colour (triple channel)
images. The CIFAR10 dataset meets this requirement in that it is a challenging
dataset of colour images. The Fashion Mnist dataset is also suitable since it a
well-tested and well understood dataset of black and white images. For reference,
the state of the art (SOA) accuracy achieved on CIFAR10 is 98.5% [22] whereas
with Fashion Mnist, the SOA accuracy is 94.6% [23].
5.1 Particle Swarm Optimization
In order to test variance and reproducibility, each configuration was run 5 times
on both CIFAR10 and Fashion Minst which resulted in the evaluation of 4000
CNN architectures for this phase of the study.
Evaluation of models trained on CIFAR10 dataset Validation accuracy
was used to evaluate the performance of both PSO configurations. It is clear from
Table 3 that the PSO model trained on swarm settings of a lower population
and higher number of iterations (population of 10 and 20 iterations) performed
significantly better. In terms of accuracy, the mean performance was 3.5% better.
Table 3. Performance of PSO models on CIFAR10
Acc Acc Acc Time
Model Layers
Max Mean StDev (min)
Population 10
0.900 0.853 0.044 1316 30
Iterations 20
Population 20
0.883 0.818 0.053 1119 38
Iterations 10
Both configurations have a very low standard deviation for model accuracy
indicating a high level of reproducibility between test runs. At a mean run time
of 21.9 hours for the first configuration and 18.6 hours for the second configura-
tion, the PSO search for CNN architectures is a slow process considering high
8 S. Lankford and D. Grimes
performance workstations, with NVIDIA GeForce GTX 1080 Ti graphic cards,
were used.
Evaluation of models trained on Fashion Mnist dataset PSO models
trained on the Fashion Mnist dataset (Table 4), achieved much higher accuracy
compared with models developed using CIFAR10 data. Similar to CIFAR10, the
low standard deviation associated with both implementations of Fashion Mnist
models indicate the PSO approach produces consistent results between different
test runs.
Table 4. Performance of PSO models on Fashion Mnist
Acc Acc Acc Time
Model Layers
Max Mean StDev (min)
Population: 10
0.943 0.932 0.009 994 30
Iterations: 20
Population: 20
0.943 0.935 0.008 1319 38
Iterations: 10
The stochastic nature of metaheuristics impacts the run times associated
with PSO for both Fashion Mnist and CIFAR10. In all tests, no clear pattern
emerged with regard to run times: CIFAR10 was faster using a population of
10 with 20 iterations whereas Fashion Mnist was faster with a population of 20
with 10 iterations. Therefore, in terms of run time, no clear conclusion could be
drawn by doubling the population and halving the iterations.
With regard to the impact of swarm settings on model accuracy for Fash-
ion Mnist, again there is little to separate the configurations. With a mean ac-
curacy of 93.5% for a population of 20 with 10 iterations and a corresponding
mean model accuracy of 93.2% using a population of 10 with 20 iterations, no
clear conclusion can be drawn.
Therefore, unlike CIFAR10, changing the swarm settings by doubling pop-
ulation and halving iterations does not impact model accuracy in the case of
Fasion Mnist. Both configurations for the PSO algorithm perform well on this
dataset.
5.2 Ant Colony Optimization
Similar to other metaheuristics, there are several parameters which can be tuned
for optimal neural architecture search using Ant Colony Optimization [17]. With
OpenNAS, users may select the options of depth, number of ants and number
of epochs in directing how the neural architecture search is conducted.
Evaluation of models trained on CIFAR10 dataset With CIFAR10 data,
the results from Table 5 indicate that a greater number of ants leads to higher
OpenNAS using SI 9
model accuracy. The improvement in max model accuracy achieved, through
doubling the number of ants and halving the number of epochs, was modest at
just 1.2%. The impact on run time for a small increase in accuracy was severe.
Doubling the number of ants effectively doubled the run time (even though the
number of epochs was halved). The standard deviation for accuracy is very low
indicating good reproducibility between the various test runs.
Table 5. Performance of ACO models on CIFAR10
Acc Acc Acc Time
Model Layers
Max Mean StDev (min)
Ants: 8
0.848 0.822 0.025 541 18
Epochs: 30
Ants: 16
0.836 0.821 0.014 1004 16
Epochs: 15
Evaluation of models trained on Fashion Mnist dataset The performance
of ACO models using Fashion Mnist data is highlighted Table 6. It can be seen
that both configurations perform well resulting in accuracies greater than 93%.
The difference in mean model accuracy between configuration A (8 ants and 30
epochs) is trivial when compared to configuration B (8 ants and 30 epochs).
However, similar to ACO on CIFAR10, the difference in run time is very signif-
icant for configuration B. Effectively it took over 7 hours longer to achieve an
accuracy improvement of 0.1%.
Clearly in the case of a simpler dataset such as Fashion Mnist, using a number
of ants in excess of 8 is not worth doing. This finding is similar to that seen with
the more complex CIFAR10 dataset, above. Therefore choosing the number of
ants, used for this ACO implementation, is an important consideration impacting
run time performance. As anticipated, the standard deviation for accuracy is also
very low indicating good reproducibility between the various test runs.
Table 6. Performance of ACO models on Fashion Mnist
Acc Acc Acc Time
Model Layers
Max Mean StDev (min)
Ants: 8
0.934 0.931 0.002 375 7
Epochs: 30
Ants: 16
0.934 0.932 0.004 837 19
Epochs: 15
10 S. Lankford and D. Grimes
Fig. 3. OpenNAS Performance of all Models on CIFAR10 and Fashion Mnist
6 Discussion
The OpenNAS performance of all models across both datasets is illustrated in
Figure 3. The results demonstrate performance comparable to that achieved by
the pso-CNN [18] approach and better than that of DeepSwarm [19].
The highest accuracy of OpenNAS in CIFAR-10 classification was 90.0%.
This was achieved using using a PSO-derived model. By comparison, DeepSwarm
achieved a top accuracy of 88.7%.
With Fashion Mnist data, the highest performing model for OpenNAS is
again a PSO derived model with an accuracy of 94.3%. This result compares
very favourably with the SOA accuracy of 94.6%. The highest performing model
for DeepSwarm achieved an accuracy of 93.56%.
In the case of pso-CNN, experiments were conducted on Fashion Mnist but
not on CIFAR-10. The best performing pso-CNN model on Fashion Mnist was
91.9% without dropout and 94.5% with dropout.
The findings clearly show that a PSO approach leads to higher model accu-
racies given that DeepSwarm is exclusively based on an ACO approach.
The pre-trained networks of MobileNet and RestNet50 delivered the poor-
est performance with CIFAR10. The other pre-trained networks, using VGG
architectures, performed very well on the same dataset.
With a more complex dataset, such as CIFAR10, the mean performance
improvement of the PSO algorithm is significant when compared with ACO.
With configurations used in this study, PSO achieved a mean accuracy of 85.3%
on CIFAR10 compared with an ACO mean accuracy of 82.2%.
The approach taken by ACO, in determining the best architecture is very
different to the PSO approach. With ACO, simpler models are initially evaluated
at lower depths with progressively more complex models being evaluated at
deeper search levels. Therefore at search depth 1, there is essentially just a
OpenNAS using SI 11
single hidden layer being evaluated. The number of ants specified creates new
architectures which simply vary the hyper parameters used for that layer. With
each new depth being explored, an additional layer is added to the architecture
being explored.
Furthermore, the ACO approach enables the targeting of hyper parameter
optimization within a given layer type rather than optimizing at the overall
architecture level. Specifying a large number of ants, with a reduced depth,
ensures the search space is restricting to studying the effects of layer hyper
parameters rather than model depth and the constituent layers. By comparison,
the number of layers in PSO generated models is entirely stochastic.
7 Conclusion
The OpenNAS approach identifies the hyperparameters within each layer of
networks used for image classification of grayscale and color datasets. In addition
the number and type of layers for the neural architecture are also identified.
This combined approach generates model architectures which achieve competitve
accuracies when classifying the CIFAR10 and Fashion Mnist datasets.
The results of swarm intelligence algorithms, in the context of this study, have
generated impressive performances. However, in many cases, their performance is
only marginally better than fine tuned pre-trained VGG models. The accuracies
of PSO derived models have been shown to exceed those of ACO derived models
in the image classification of grayscale and color datasets.
In addition, the OpenNAS integrated approach, using both PSO and ACO al-
gorithms, yields higher accuracies when compared with DeepSwarm which relies
on a single metaheuristic.
References
1. F. Hutter, L. Kotthoff, and J. Vanschoren, Automated machine learning: methods,
systems, challenges. Springer Nature, 2019.
2. L. Kotthoff, C. Thornton, H. H. Hoos, F. Hutter, and K. Leyton-Brown, “Auto-
weka 2.0: Automatic model selection and hyperparameter optimization in weka,”
The Journal of Machine Learning Research, vol. 18, no. 1, pp. 826–830, 2017.
3. B. Komer, J. Bergstra, and C. Eliasmith, “Hyperopt-sklearn,” in Automated Ma-
chine Learning. Springer, Cham, 2019, pp. 97–111.
4. H. Jin, Q. Song, and X. Hu, “Auto-keras: An efficient neural architecture search
system,” in Proceedings of the 25th ACM SIGKDD International Conference on
Knowledge Discovery & Data Mining, 2019, pp. 1946–1956.
5. M. Feurer, A. Klein, K. Eggensperger, J. T. Springenberg, M. Blum, and F. Hutter,
“Auto-sklearn: efficient and robust automated machine learning,” in Automated
Machine Learning. Springer, Cham, 2019, pp. 113–134.
6. M. Feurer, K. Eggensperger, S. Falkner, M. Lindauer, and F. Hutter, “Auto-sklearn
2.0: The next generation,” arXiv preprint arXiv:2007.04074, 2020.
7. R. S. Olson and J. H. Moore, “Tpot: A tree-based pipeline optimization tool for
automating,” Automated Machine Learning: Methods, Systems, Challenges, p. 151,
2019.
12 S. Lankford and D. Grimes
8. B. A. Garro and R. A. Vázquez, “Designing artificial neural networks using particle
swarm optimization algorithms,” Computational intelligence and neuroscience, vol.
2015, 2015.
9. M. Mavrovouniotis and S. Yang, “Training neural networks with ant colony opti-
mization algorithms for pattern classification,” Soft Computing, vol. 19, no. 6, pp.
1511–1522, 2015.
10. Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied
to document recognition,” Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324,
1998.
11. K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recog-
nition,” in Proceedings of the IEEE conference on computer vision and pattern
recognition, 2016, pp. 770–778.
12. K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale
image recognition,” arXiv preprint arXiv:1409.1556, 2014.
13. T. Elsken, J. H. Metzen, F. Hutter et al., “Neural architecture search.” 2019.
14. V. Sze, Y.-H. Chen, T.-J. Yang, and J. S. Emer, “Efficient processing of deep neural
networks: A tutorial and survey,” Proceedings of the IEEE, vol. 105, no. 12, pp.
2295–2329, 2017.
15. J. Kennedy and R. Eberhart, “Particle swarm optimization,” in Proceedings of
ICNN’95-International Conference on Neural Networks, vol. 4. IEEE, 1995, pp.
1942–1948.
16. R. C. Eberhart and Y. Shi, “Comparison between genetic algorithms and particle
swarm optimization,” in International conference on evolutionary programming.
Springer, 1998, pp. 611–616.
17. M. Dorigo and L. M. Gambardella, “Ant colony system: a cooperative learning
approach to the traveling salesman problem,” IEEE Transactions on evolutionary
computation, vol. 1, no. 1, pp. 53–66, 1997.
18. F. E. F. Junior and G. G. Yen, “Particle swarm optimization of deep neural net-
works architectures for image classification,” Swarm and Evolutionary Computa-
tion, vol. 49, pp. 62–74, 2019.
19. E. Byla and W. Pang, “Deepswarm: Optimising convolutional neural networks us-
ing swarm intelligence,” in UK Workshop on Computational Intelligence. Springer,
2019, pp. 119–130.
20. A. Krizhevsky, V. Nair, and G. Hinton, “The cifar-10 dataset,” online: http://www.
cs. toronto. edu/kriz/cifar. html, vol. 55, 2014.
21. H. Xiao, K. Rasul, and R. Vollgraf, “Fashion-mnist: a novel image dataset for
benchmarking machine learning algorithms,” arXiv preprint arXiv:1708.07747,
2017.
22. E. D. Cubuk, B. Zoph, D. Mane, V. Vasudevan, and Q. V. Le, “Autoaugment:
Learning augmentation strategies from data,” in Proceedings of the IEEE confer-
ence on computer vision and pattern recognition, 2019, pp. 113–123.
23. B. Ma, X. Li, Y. Xia, and Y. Zhang, “Autonomous deep learning: A genetic dcnn
designer for image classification,” Neurocomputing, vol. 379, pp. 152–161, 2020.