Literature on Neural Architecture Search

@article{BALDEONCALISTO2021102154,

title = {EMONAS-Net: Efficient multiobjective neural architecture search using surrogate-assisted evolutionary algorithm for 3D medical image segmentation},

author = {Maria Baldeon Calisto and Susana K. Lai-Yuen},

url = {https://www.sciencedirect.com/science/article/pii/S0933365721001470},

doi = {https://doi.org/10.1016/j.artmed.2021.102154},

issn = {0933-3657},

year  = {2021},

date = {2021-01-01},

urldate = {2021-01-01},

journal = {Artificial Intelligence in Medicine},

volume = {119},

pages = {102154},

abstract = {Deep learning plays a critical role in medical image segmentation. Nevertheless, manually designing a neural network for a specific segmentation problem is a very difficult and time-consuming task due to the massive hyperparameter search space, long training time and large volumetric data. Therefore, most designed networks are highly complex, task specific and over-parametrized. Recently, multiobjective neural architecture search (NAS) methods have been proposed to automate the design of accurate and efficient segmentation architectures. However, they only search for either the micro- or macro-structure of the architecture, do not use the information produced during the optimization process to increase the efficiency of the search, or do not consider the volumetric nature of medical images. In this work, we present EMONAS-Net, an Efficient MultiObjective NAS framework for 3D medical image segmentation that optimizes both the segmentation accuracy and size of the network. EMONAS-Net has two key components, a novel search space that considers the configuration of the micro- and macro-structure of the architecture and a Surrogate-assisted Multiobjective Evolutionary based Algorithm (SaMEA algorithm) that efficiently searches for the best hyperparameter values. The SaMEA algorithm uses the information collected during the initial generations of the evolutionary process to identify the most promising subproblems and select the best performing hyperparameter values during mutation to improve the convergence speed. Furthermore, a Random Forest surrogate model is incorporated to accelerate the fitness evaluation of the candidate architectures. EMONAS-Net is tested on the tasks of prostate segmentation from the MICCAI PROMISE12 challenge, hippocampus segmentation from the Medical Segmentation Decathlon challenge, and cardiac segmentation from the MICCAI ACDC challenge. In all the benchmarks, the proposed framework finds architectures that perform better or comparable with competing state-of-the-art NAS methods while being considerably smaller and reducing the architecture search time by more than 50%.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{FAN202137,

title = {Determining learning direction via multi-controller model for stably searching generative adversarial networks},

author = {Yi Fan and Quoqiang Zhou and Weifeng Zhang and Shudi Bao and Jun Shen},

url = {https://www.sciencedirect.com/science/article/pii/S0925231221012601},

doi = {https://doi.org/10.1016/j.neucom.2021.08.070},

issn = {0925-2312},

year  = {2021},

date = {2021-01-01},

urldate = {2021-01-01},

journal = {Neurocomputing},

volume = {464},

pages = {37-47},

abstract = {The data generated by Generative Adversarial Network (GAN) inevitably contains noise, which can be reduced by searching and optimizing the architecture of GAN. To search for generative adversarial networks architectures stably, a neural architecture search (NAS) method, StableAutoGAN, is proposed based on the existing algorithm, AutoGAN. The stability of conventional reinforcement learning (RL)-based NAS methods for GAN is adversely influenced by the uncertainty of direction, where the controller will go forward once receiving inaccurate rewards. In StableAutoGAN, a multi-controller model is employed to mitigate this problem via comparing the performance of controllers after receiving rewards. During the search process, each controller independently learns the sampling policy. Meanwhile, the learning effect is measured by the credibility score, which further determines the usage of controllers. Our experiments show that the standard deviation of Frchet Inception Distance (FID) scores of the GANs discovered by StableAutoGAN is approximately 1/16 and 1/8 of that by AutoGAN on CIFAR-10 and on STL-10 respectively, while the effects remain similar to AutoGAN.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{9518382,

title = {Neural Architecture Search for Portrait Parsing},

author = {Bo Lyu and Yin Yang and Shiping Wen and Tingwen Huang and Ke Li},

url = {https://ieeexplore.ieee.org/abstract/document/9518382},

doi = {10.1109/TNNLS.2021.3104872},

year  = {2021},

date = {2021-01-01},

urldate = {2021-01-01},

journal = {IEEE Transactions on Neural Networks and Learning Systems},

pages = {1-10},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{DBLP:journals/information/RazaliMYHHMH21,

title = {Indigenous Food Recognition Model Based on Various Convolutional Neural  Network Architectures for Gastronomic Tourism Business Analytics},

author = {Mohd Norhisham Razali and Ervin Gubin Moung and Farashazillah Yahya and Chong Joon Hou and Rozita Hanapi and Raihani Mohamed and Ibrahim Abakr Targio Hashem},

url = {https://doi.org/10.3390/info12080322},

doi = {10.3390/info12080322},

year  = {2021},

date = {2021-01-01},

urldate = {2021-01-01},

journal = {Inf.},

volume = {12},

number = {8},

pages = {322},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{DBLP:journals/corr/abs-2108-05647,

title = {DARTS for Inverse Problems: a Study on Hyperparameter Sensitivity},

author = {Jonas Geiping and Jovita Lukasik and Margret Keuper and Michael Moeller},

url = {https://arxiv.org/abs/2108.05647},

year  = {2021},

date = {2021-01-01},

urldate = {2021-01-01},

journal = {CoRR},

volume = {abs/2108.05647},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{9513550,

title = {Random Topology and Random Multiscale Mapping: An Automated Design of Multiscale and Lightweight Neural Network for Remote-Sensing Image Recognition},

author = {Jihao Li and Martin Weinmann and Xian Sun and Wenhui Diao and Yingchao Feng and Kun Fu},

url = {https://ieeexplore.ieee.org/abstract/document/9513550},

doi = {10.1109/TGRS.2021.3102988},

year  = {2021},

date = {2021-01-01},

urldate = {2021-01-01},

journal = {IEEE Transactions on Geoscience and Remote Sensing},

pages = {1-17},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{REPIN2021104788,

title = {Automated Deep Abstractions for Stochastic Chemical Reaction Networks},

author = {Denis Repin and Tatjana Petrov},

url = {https://www.sciencedirect.com/science/article/pii/S0890540121001048},

doi = {https://doi.org/10.1016/j.ic.2021.104788},

issn = {0890-5401},

year  = {2021},

date = {2021-01-01},

journal = {Information and Computation},

pages = {104788},

abstract = {Predicting stochastic cellular dynamics as emerging from the mechanistic models of molecular interactions is a long-standing challenge in systems biology: low-level chemical reaction network (CRN) models give rise to a highly-dimensional continuous-time Markov chain (CTMC) which is computationally demanding and often prohibitive to analyse in practice. A recently proposed abstraction method uses deep learning to replace this CTMC with a discrete-time continuous-space process, by training a mixture density deep neural network with traces sampled at regular time intervals (which can be obtained either by simulating a given CRN or as time-series data from experiment). The major advantage of such abstraction is that it produces a computational model that is dramatically cheaper to execute, while it preserves the statistical features of the training data. In general, the abstraction accuracy improves with the amount of training data. However, depending on the CRN, the overall quality of the method – the efficiency gain and abstraction accuracy – will also depend on the choice of neural network architecture given by hyper-parameters such as the layer types and connections between them. As a consequence, in practice, the modeller has to take care of finding the suitable architecture manually, for each given CRN, through a tedious and time-consuming trial-and-error cycle. In this paper, we propose to further automatise deep abstractions for stochastic CRNs, through learning the neural network architecture along with learning the transition kernel of the abstract process. Automated search of the architecture makes the method applicable directly to any given CRN, which is time-saving for deep learning experts and crucial for non-specialists. We implement the method and demonstrate its performance on a number of representative CRNs with multi-modal emergent phenotypes. Moreover, we showcase that deep abstractions can be used for efficient multi-scale simulations, which are otherwise computationally intractable. To this end, we define a scenario where multiple CRN instances interact across a spatial grid via shared species. Finally, we discuss the limitations and challenges arising when using deep abstractions.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{9508863,

title = {Evolutional Neural Architecture Search for Optimization of Spatiotemporal Brain Network Decomposition},

author = {Qing Li and Wei Zhang and Lin Zhao and Xia Wu and Tianming Liu},

url = {https://ieeexplore.ieee.org/abstract/document/9508863},

doi = {10.1109/TBME.2021.3102466},

year  = {2021},

date = {2021-01-01},

journal = {IEEE Transactions on Biomedical Engineering},

pages = {1-1},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{HE2021100955,

title = {Simple-Encoded Evolving Convolutional Neural Networks and Its Application to Skin Disease Image Classification},

author = {Xiaoyu He and Yong Wang and Xiaojing Wang and Weihong Huang and Shuang Zhao and Xiang Chen},

url = {https://www.sciencedirect.com/science/article/pii/S2210650221001176},

doi = {https://doi.org/10.1016/j.swevo.2021.100955},

issn = {2210-6502},

year  = {2021},

date = {2021-01-01},

journal = {Swarm and Evolutionary Computation},

pages = {100955},

abstract = {Automatically searched architectures by neural architecture search (NAS) methods have shown promising performance in various visual recognition tasks. Among NAS methods, evolutionary neural architecture methods are popular because of their potential to find the global optimal convolutional neural networks (CNNs). These methods usually use an individual to represent a CNN, while often facing two challenges: 1) since a CNN has numerous encoded parameters and weights, the length of each individual is long which causes a large search space, and 2) due to the unknown optimal depth of a CNN, it is necessary to deal with a variable-length optimization problem, which leads to the search in a messy way since search spaces with different dimensions may have different optimal solutions. In this paper, we propose a genetic algorithm with a simple encoding scheme (SEECNN) for evolving CNNs to address image classification problems. In our encoding scheme, the parameters and weights of each layer are encoded into an individual and the whole population represents an entire CNN. Over the course of evolution, three offspring subpopulations are separately produced by genetic operators on three subpopulations. In each subpopulation, the lengths of individuals are short and equal. Afterward, we design a stable search strategy to update the population based on the performance improvement, where we only insert, replace, and remove one individual to generate candidate populations. SEECNN is compared with 24 well-known algorithms on nine benchmark datasets and five state-of-the-art methods on a real-world skin disease image classification case. The results demonstrate its effectiveness.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{9503404,

title = {Exploring Neural Architecture Search Space via Deep Deterministic Sampling},

author = {Keith G Mills and Mohammad Salameh and Di Niu and Fred X Han and Seyed Saeed Changiz Rezaei and Hengshuai Yao and Wei Lu and Shuo Lian and Shangling Jui},

url = {https://ieeexplore.ieee.org/abstract/document/9503404},

doi = {10.1109/ACCESS.2021.3101975},

year  = {2021},

date = {2021-01-01},

journal = {IEEE Access},

volume = {9},

pages = {110962-110974},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{9497513,

title = {Searching for CNN Architectures for Remote Sensing Scene Classification},

author = {Clifford Broni-Bediako and Yuki Murata and Luiz H B Mormille and Masayasu Atsumi},

doi = {10.1109/TGRS.2021.3097938},

year  = {2021},

date = {2021-01-01},

journal = {IEEE Transactions on Geoscience and Remote Sensing},

pages = {1-13},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{DBLP:journals/corr/abs-2107-13153,

title = {Homogeneous Architecture Augmentation for Neural Predictor},

author = {Yuqiao Liu and Yehui Tang and Yanan Sun},

url = {https://arxiv.org/abs/2107.13153},

year  = {2021},

date = {2021-01-01},

journal = {CoRR},

volume = {abs/2107.13153},

keywords = {},

pubstate = {published},

tppubtype = {article}

}