RNAS-CL: Robust Neural Architecture Search by Cross-Layer Knowledge Distillation.

ETNAS: An energy consumption task-driven neural architecture search

Deep learning has made substantial breakthroughs in many fields due to its powerful automatic representation capabilities. It has been proven that neural architecture design is crucial to the feature representation of data and the final performance. However, the design of the neural architecture heavily relies on the researchers’ prior knowledge and experience. And due to the limitations of humans’ inherent knowledge, it is difficult for people to jump out of their original thinking paradigm and design an optimal model. Therefore, an intuitive idea would be to reduce human intervention as much as possible and let the algorithm automatically design the neural architecture.Neural Architecture Search(NAS) is just such a revolutionary algorithm, and the related research work is complicated and rich. Therefore, a comprehensive and systematic survey on the NAS is essential. Previously related surveys have begun to classify existing work mainly based on the key components of NAS: search space, search strategy, and evaluation strategy. While this classification method is more intuitive, it is difficult for readers to grasp the challenges and the landmark work involved. Therefore, in this survey, we provide a new perspective: beginning with an overview of the characteristics of the earliest NAS algorithms, summarizing the problems in these early NAS algorithms, and then providing solutions for subsequent related research work. In addition, we conduct a detailed and comprehensive analysis, comparison, and summary of these works. Finally, we provide some possible future research directions.

Neural architecture search (NAS) targets at finding the optimal architecture of a neural network for a problem or a family of problems. Evaluations of neural architectures are very time-consuming. One of the possible ways to mitigate this issue is to use low-fidelity evaluations, namely training on a part of a dataset, fewer epochs, with fewer channels, etc. In this paper, we propose a Bayesian multi-fidelity method for neural architecture search: MF-KD. The method relies on a new approach to low-fidelity evaluations of neural architectures by training for a few epochs using a knowledge distillation. Knowledge distillation adds to a loss function a term forcing a network to mimic some teacher network. We carry out experiments on CIFAR-10, CIFAR-100, and ImageNet-16-120. We show that training for a few epochs with such a modified loss function leads to a better selection of neural architectures than training for a few epochs with a logistic loss. The proposed method outperforms several state-of-the-art baselines.

Decision of the model complexity is a significant challenge in contemporary data-driven modeling applications. Designing neural architecture involves the process of determining the optimal model complexity for deep neural networks (DNNs) models in order to uncover relationships in real-world data patterns. Consequently, optimizing DNN architectures is crucial for enhancing the practical approximation performance of DNN models to real-world data. This study implements a data-driven neuroevolution scheme for the optimal neural architecture search (NAS) and demonstrates a data-driven engineering application for cooling load estimation. The proposed neuroevolution scheme aims at evolving to the best generalizing DNN model that well suits the modeling complexity requirements of the dataset. To this end, the objective function for the neural architecture optimization process is simplified to the mean square error of the test dataset, which enables to reduce the risk of insufficient generalization during NAS. By employing this objective function for evolution field optimization (EFO), the proposed neuroevolution process can automatically achieve the optimal model complexity, preventing overfitting and underfitting cases, and thereby attaining almost the best generalization for the dataset. For this purpose, this approach combines parametric learning with the backpropagation algorithm and structural learning with EFO-based neural architecture search to address data-driven, optimal complexity DNN model generation problems. Effectiveness of the method is demonstrated in the cooling load estimation problem of residential buildings, and performances of the optimal DNN models with four objective functions are analyzed. The design of objective function for the best generalizing model is also elaborated.

Anomalous actors are becoming increasingly sophisticated both in methodology and technical ability. Fortunately, the companies impacted by anomalous behaviour are also generating more data and insights on potential anomalous cases than ever before. In this paper, a unified approach to managing the complexity of constructing a useable anomaly detection system is presented. The unified approach is comprised of three algorithms, a Neural Architecture Search (NAS) implementation for autoencoders, an anomaly score threshold optimisation algorithm, and a Gaussian scaling function for anomaly scores. NAS is applied to a data set containing instances of credit card fraud. The NAS algorithm is used to simulate a population of 50 candidate deep learning architectures, with the best performing architecture being selected based on a balanced score, comprised of an average of the Area under the ROC curve, Average Precision and normalised Matthews Correlation Coefficient scores. The threshold optimisation algorithm is used to determine the appropriate threshold between the classes, for the purposes of producing the binary classification outcome of each architecture. The Gaussian scaling algorithm is applied to the raw anomaly scores of the optimal architecture into order to generate useable probability scores. Not only did the proposed unified approach simplify the process of selecting an optimal neural architecture whose output is interpretable by business practitioners and comparable with other probability score producing models, but it also contributes to anomaly detection in a transactional setting by eliminating subjective thresholds when classifying anomalous transactions.

In the future, vehicles will be equipped with increasingly advanced interactive intelligent electronic devices, which will induce drivers to conduct secondary tasks, thereby leading to distractions. Therefore, the detection and early warning of driver distraction are essential for improving driving safety and pose an important challenge in intelligent transportation systems. Previous studies used traditional machine learning and deep learning transfer models, which have the disadvantages of complicated and time-consuming manual feature engineering, strong subjectivity, and weak generalization performance. In this paper, we propose a fine-grained detection method for driver distraction based on neural architecture search. First, we design an automatic construction algorithm for deep convolutional neural networks based on neural architecture search, which automatically searches for the optimal deep convolutional neural network architecture without human involvement. In addition, we fuse driver-related multisource perception information, use an automatically constructed deep convolutional neural network to extract high-dimensional mapping features, and implement fine-grained detection of various types of driver distraction states. The results on a large-scale multimodal driver distraction dataset demonstrate that our method can efficiently search an optimal deep convolutional neural network, which can quickly converge, and can accurately detect the considered types of driver distraction states, the average detection accuracy reaches 99.7796%; moreover, it has satisfactory robustness.

Recent breakthroughs of Neural Architecture Search (NAS) extend the field’s research scope towards a broader range of vision tasks and more diversified search spaces. While existing NAS methods mostly design architectures on a single task, algorithms that look beyond single-task search are surging to pursue a more efficient and universal solution across various tasks. Many of them leverage transfer learning and seek to preserve, reuse, and refine network design knowledge to achieve higher efficiency in future tasks. However, the enormous computational cost and experiment complexity of cross-task NAS are imposing barriers for valuable research in this direction. Existing NAS benchmarks all focus on one type of vision task, i.e., classification. In this work, we propose TransNAS-Bench-101, a benchmark dataset containing network performance across seven tasks, covering classification, regression, pixel-level prediction, and self-supervised tasks. This diversity provides opportunities to transfer NAS methods among tasks and allows for more complex transfer schemes to evolve. We explore two fundamentally different types of search space: cell-level search space and macro-level search space. With 7,352 backbones evaluated on seven tasks, 51,464 trained models with detailed training information are provided. With TransNAS-Bench-101, we hope to encourage the advent of exceptional NAS algorithms that raise cross-task search efficiency and generalizability to the next level. Our dataset and code will be available at Mindspore <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> and VEGA <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> .

Improving the performance of deep neural networks (DNNs) is important to both the compiler and neural architecture search (NAS) communities. Compilers apply program transformations in order to exploit hardware parallelism and memory hierarchy. However, legality concerns mean they fail to exploit the natural robustness of neural networks. In contrast, NAS techniques mutate networks by operations such as the grouping or bottlenecking of convolutions, exploiting the resilience of DNNs. In this work, we express such neural architecture operations as program transformations whose legality depends on a notion of representational capacity. This allows them to be combined with existing transformations into a unified optimization framework. This unification allows us to express existing NAS operations as combinations of simpler transformations. Crucially, it allows us to generate and explore new tensor convolutions. We prototyped the combined framework in TVM and were able to find optimizations across different DNNs, that significantly reduce inference time - over 3× in the majority of cases. Furthermore, our scheme dramatically reduces NAS search time.

POPNASv3: A pareto-optimal neural architecture search solution for image and time series classification

With the emergence of deep neural networks, many research fields, such as image classification, object detection, speech recognition, natural language processing, machine translation and automatic driving, have made major breakthroughs in technology and the research achievements have been successfully applied in many real-life applications. Combining evolutionary computation and neural architecture search (NAS) is an important approach to improve the performance of deep neural networks. Usually, the related researchers only focus on precision. Thus, the searched neural architectures always perform poorly in the other indexes such as time cost. In this paper, a multi-objective evolutionary algorithm with a probability stack (MOEA-PS) is proposed for NAS, which considers the two objects of precision and time consumption. MOEA-PS uses an adjacency list to represent the internal structure of deep neural networks. Besides, a unique mechanism is introduced into the multi-objective genetic algorithm to guide the process of crossover and mutation when generating offspring. Furthermore, the structure blocks are stacked using a proxy model to generate deep neural networks. The results of the experiments on Cifar-10 and Cifar-100 demonstrate that the proposed algorithm has a similar error rate compared with the most advanced NAS algorithms, but the time cost is lower. Finally, the network structure searched on Cifar-10 is transferred directly to the ImageNet dataset, which can achieve 73.6% classification accuracy.

Neural Architecture Search (NAS) is a busy research field growing exponentially in recent years. State-of-the-art deep neural networks usually require a specialist to fine-tune the model to solve a specific problem. NAS research aims to design neural network architectures automatically, thus easing the need for machine learning specialists to spend a lot of effort on hand-crafted attempts. As artificial intelligence applications are becoming ubiquitous, there is also a growing interest in efficient applications that could be deployed to smartphones, smart wearable devices, and other edge devices. Gender recognition in unfiltered images — such as those we find in real-world situations like pictures taken with smartphones and video shots from surveillance cameras — is one of such challenging applications. In this work, we developed an evolutionary NAS algorithm that consistently finds efficient ResNet-based architectures, named RENNAS, which have a good trade-off between classification accuracy and architectural and computational complexities. We demonstrate our algorithm's performance on Adience dataset of unfiltered images for gender recognition.

Recent AI research has witnessed increasing interests in automatically designing the architecture of deep neural networks, which is coined as neural architecture search (NAS). The automatically searched network architectures via NAS methods have outperformed manually designed architectures on some NLP tasks. However, training a large number of model configurations for efficient NAS is computationally expensive, creating a substantial barrier for applying NAS methods in real-life applications. In this paper, we propose to accelerate neural architecture search for natural language processing based on knowledge distillation (called KD-NAS). Specifically, instead of searching the optimal network architecture on the validation set conditioned on the optimal network weights on the training set, we learn the optimal network by minimizing the knowledge loss transferred from a pre-trained teacher network to the searching network based on Earth Mover's Distance (EMD). Experiments on five datasets show that our method achieves promising performance compared to strong competitors on both accuracy and searching speed. For reproducibility, we submit the code at: https://github.com/lxk00/KD-NAS-EMD.

Parameter-less Pareto local search for multi-objective neural architecture search with the Interleaved Multi-start Scheme

A surrogate evolutionary neural architecture search algorithm for graph neural networks

Neural architecture search (NAS) is a promising approach to fully automated machine learning that applies to arbitrary tasks. However, existing NAS methods tend to make a complex network that requires much time to train and predict. Since machine learning applications are now often run on mobile devices with limited computational resources, there is a need to obtain small, efficient, yet effective network models. This paper proposes MIA-NAS (Mutual Information Analysis Neural Architecture Search), an evolutionary NAS algorithm that extends Neural Architecture Search by Hill Climbing (NASH) by Elsken et al.. The proposed method introduces a procedure that removes layers that are less contributing to making predictions. The amount of contribution from each layer is quantified by estimating mutual information they have with true labels. The experiments showed that MIA-NAS produces smaller networks with comparative performance to the ones made by NASH. Besides, removing layers based on mutual information resulted in better performance than randomly selecting layers to be removed.