Automated Safety Verification of Programs Invoking Neural Networks

AbstractState-of-the-art program-analysis techniques are not yet able to effectively verify safety properties of heterogeneous systems, that is, systems with components implemented using diverse technologies. This shortcoming is pinpointed by programs invoking neural networks despite their acclaimed role as innovation drivers across many application areas. In this paper, we embark on the verification of system-level properties for systems characterized by interaction between programs and neural networks. Our technique provides a tight two-way integration of a program and a neural-network analysis and is formalized in a general framework based on abstract interpretation. We evaluate its effectiveness on 26 variants of a widely used, restricted autonomous-driving benchmark.

Download Full-text

Deep neural network analysis of nanoparticle ordering to identify defects in layered carbon materials

Chemical Science ◽

10.1039/d0sc05696k ◽

2021 ◽

Author(s):

Daniil A. Boiko ◽

Evgeniy O. Pentsak ◽

Vera A. Cherepanova ◽

Evgeniy G. Gordeev ◽

Valentine P. Ananikov

Keyword(s):

Neural Network ◽

Neural Networks ◽

Network Analysis ◽

Carbon Material ◽

Deep Neural Network ◽

Carbon Materials ◽

Material Surface ◽

Neural Network Analysis ◽

Pd Nanoparticle ◽

Sem Imaging

Defectiveness of carbon material surface is a key issue for many applications. Pd-nanoparticle SEM imaging was used to highlight “hidden” defects and analyzed by neural networks to solve order/disorder classification and defect segmentation tasks.

Download Full-text

Quantitative Dynamics Effects of Belt and Road Economies Trade Using Structural Gravity and Neural Networks

SAGE Open ◽

10.1177/21582440211032662 ◽

2021 ◽

Vol 11 (3) ◽

pp. 215824402110326

Author(s):

Koffi Dumor ◽

Li Yao ◽

Jean-Paul Ainam ◽

Edem Koffi Amouzou ◽

Williams Ayivi

Keyword(s):

Neural Network ◽

Neural Networks ◽

Network Analysis ◽

Bilateral Trade ◽

Belt And Road Initiative ◽

Neural Network Analysis ◽

Analysis Techniques ◽

Belt And Road ◽

The Impact ◽

Trade Partners

Recent research suggests that China’s Belt and Road Initiative (BRI) would improve the bilateral trade between China and its partners. This article uses detailed bilateral export data from 1990 to 2017 to investigate the impact of China’s BRI on its trade partners using neural network analysis techniques and structural gravity model estimations. Our main findings suggest that the BRI countries would raise exports by a modest 5.053%. This indicates that export and network upgrades should be considered from economic and policy perspectives. The results also show that neural networks is more robust compared with structural gravity framework.

Download Full-text

SADA: Semantic Adversarial Diagnostic Attacks for Autonomous Applications

Proceedings of the AAAI Conference on Artificial Intelligence ◽

10.1609/aaai.v34i07.6722 ◽

2020 ◽

Vol 34 (07) ◽

pp. 10901-10908 ◽

Cited By ~ 2

Author(s):

Abdullah Hamdi ◽

Matthias Mueller ◽

Bernard Ghanem

Keyword(s):

Neural Networks ◽

Recent Work ◽

Autonomous Navigation ◽

General Framework ◽

Deep Neural Networks ◽

Autonomous Driving ◽

Black Box ◽

Semantic Meaning ◽

Safety Critical ◽

Adversarial Attack

One major factor impeding more widespread adoption of deep neural networks (DNNs) is their lack of robustness, which is essential for safety-critical applications such as autonomous driving. This has motivated much recent work on adversarial attacks for DNNs, which mostly focus on pixel-level perturbations void of semantic meaning. In contrast, we present a general framework for adversarial attacks on trained agents, which covers semantic perturbations to the environment of the agent performing the task as well as pixel-level attacks. To do this, we re-frame the adversarial attack problem as learning a distribution of parameters that always fools the agent. In the semantic case, our proposed adversary (denoted as BBGAN) is trained to sample parameters that describe the environment with which the black-box agent interacts, such that the agent performs its dedicated task poorly in this environment. We apply BBGAN on three different tasks, primarily targeting aspects of autonomous navigation: object detection, self-driving, and autonomous UAV racing. On these tasks, BBGAN can generate failure cases that consistently fool a trained agent.

Download Full-text

Estimating China’s Trade with Its Partner Countries within the Belt and Road Initiative Using Neural Network Analysis

Sustainability ◽

10.3390/su11051449 ◽

2019 ◽

Vol 11 (5) ◽

pp. 1449 ◽

Cited By ~ 1

Author(s):

Koffi Dumor ◽

Li Yao

Keyword(s):

Neural Network ◽

Neural Networks ◽

International Trade ◽

Network Analysis ◽

Gravity Model ◽

Accurate Estimation ◽

Belt And Road Initiative ◽

Neural Network Analysis ◽

Belt And Road ◽

The Belt And Road

The Belt and Road Initiative (BRI) under the auspices of the Chinese government was created as a regional integration and development model between China and her trade partners. Arguments have been raised as to whether this initiative will be beneficial to participating countries in the long run. We set to examine how to estimate this trade initiative by comparing the relative estimation powers of the traditional gravity model with the neural network analysis using detailed bilateral trade exports data from 1990 to 2017. The results show that neural networks are better than the gravity model approach in learning and clarifying international trade estimation. The neural networks with fixed country effects showed a more accurate estimation compared to a baseline model with country-year fixed effects, as in the OLS estimator and Poisson pseudo-maximum likelihood. On the other hand, the analysis indicated that more than 50% of the 6 participating East African countries in the BRI were able to attain their predicted targets. Kenya achieved an 80% (4 of 5) target. Drawing from the lessons of the BRI and the use of neural network model, it will serve as an important reference point by which other international trade interventions could be measured and compared.

Download Full-text

Mechanisms of Modulation of a Neural Network

Physiology ◽

10.1152/physiologyonline.2000.15.4.186 ◽

2000 ◽

Vol 15 (4) ◽

pp. 186-191 ◽

Cited By ~ 3

Author(s):

Abdeljabbar El Manira ◽

Peter Wallén

Keyword(s):

Neural Network ◽

Neural Networks ◽

Network Analysis ◽

Ion Channel ◽

Patterns Of Behavior ◽

And Control

Neural networks form the basis for the generation and control of various patterns of behavior. Such networks are subjected to modulatory systems that influence their operation and, thereby, the behavior. In the lamprey locomotor network, analysis on the ion channel, synaptic, and cellular levels has given new insights into the organization of such modulatory systems.

Download Full-text

Implementation, Algorithmic, and Computational Level Descriptions of Neural Networks. Review ofMathematical Methods for Neural Network Analysis and Design,by R. M. Golden

Journal of Mathematical Psychology ◽

10.1006/jmps.1997.1195 ◽

1998 ◽

Vol 42 (1) ◽

pp. 107-109

Author(s):

Jerome R. Busemeyer

Keyword(s):

Neural Network ◽

Neural Networks ◽

Network Analysis ◽

Neural Network Analysis ◽

Analysis And Design ◽

Computational Level

Download Full-text

Bankruptcy in Indian Private Sector Banks: A Neural Network Analysis

Ushus - Journal of Business Management ◽

10.12725/ujbm.52.6 ◽

2020 ◽

Vol 19 (3) ◽

pp. 89-114

Author(s):

Surbhi Dhama

Keyword(s):

Neural Network ◽

Neural Networks ◽

Network Analysis ◽

Private Sector ◽

Financial Institutions ◽

Economic Condition ◽

Banking Industry ◽

Financial Data ◽

Neural Network Analysis ◽

Private Banks

This paper aims to predict the bankruptcy in Indian private banks using financial ratios such as ROA, GNPA, EPS, PAT, and GNP of the country. This paper also explains the importance of Ohlson’s number, Graham’s number and Zmijewski number as the major predictors of bankruptcy while developing a model using neural networks. For the prediction, the financial data for private sector banks of India such as HDFC, HDFC, ICICI, AXIS, YES bank, KOTAK MAHINDRA Bank, FEDERAL BANK, INDUSIND Bank, RBL and KARUR VYSYA for the last 10 years from 2010-2019 have been analysed. The model developed during the research will help the financial institutions and banks in India to understand the economic condition of the banking industry.

Download Full-text

Lightweight Convolutional Neural Networks with Model-Switching Architecture for Multi-Scenario Road Semantic Segmentation

Applied Sciences ◽

10.3390/app11167424 ◽

2021 ◽

Vol 11 (16) ◽

pp. 7424

Author(s):

Peng-Wei Lin ◽

Chih-Ming Hsu

Keyword(s):

Neural Network ◽

Neural Networks ◽

Convolutional Neural Network ◽

Semantic Segmentation ◽

Autonomous Driving ◽

Individual Model ◽

Suppression Effect ◽

Optimal Weights ◽

Model Size ◽

Multiple Scenarios

A convolutional neural network (CNN) that was trained using datasets for multiple scenarios was proposed to facilitate real-time road semantic segmentation for various scenarios encountered in autonomous driving. However, the CNN inhibited the mutual suppression effect between weights; thus, it did not perform as well as a network that was trained using a single scenario. To address this limitation, we used a model-switching architecture in the network and maintained the optimal weights of each individual model which required considerable space and computation. We, subsequently, incorporated a lightweight process into the model to reduce the model size and computational load. The experimental results indicated that the proposed lightweight CNN with a model-switching architecture outperformed and was faster than the conventional methods across multiple scenarios in road semantic segmentation.

Download Full-text

Optical coherent dot-product chip for sophisticated deep learning regression

Light Science & Applications ◽

10.1038/s41377-021-00666-8 ◽

2021 ◽

Vol 10 (1) ◽

Author(s):

Shaofu Xu ◽

Jing Wang ◽

Haowen Shu ◽

Zhike Zhang ◽

Sicheng Yi ◽

...

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Language Processing ◽

High Speed ◽

Autonomous Driving ◽

Single Chip ◽

Optical Fields ◽

Large Bandwidth ◽

Dot Product

AbstractOptical implementations of neural networks (ONNs) herald the next-generation high-speed and energy-efficient deep learning computing by harnessing the technical advantages of large bandwidth and high parallelism of optics. However, due to the problems of the incomplete numerical domain, limited hardware scale, or inadequate numerical accuracy, the majority of existing ONNs were studied for basic classification tasks. Given that regression is a fundamental form of deep learning and accounts for a large part of current artificial intelligence applications, it is necessary to master deep learning regression for further development and deployment of ONNs. Here, we demonstrate a silicon-based optical coherent dot-product chip (OCDC) capable of completing deep learning regression tasks. The OCDC adopts optical fields to carry out operations in the complete real-value domain instead of in only the positive domain. Via reusing, a single chip conducts matrix multiplications and convolutions in neural networks of any complexity. Also, hardware deviations are compensated via in-situ backpropagation control provided the simplicity of chip architecture. Therefore, the OCDC meets the requirements for sophisticated regression tasks and we successfully demonstrate a representative neural network, the AUTOMAP (a cutting-edge neural network model for image reconstruction). The quality of reconstructed images by the OCDC and a 32-bit digital computer is comparable. To the best of our knowledge, there is no precedent of performing such state-of-the-art regression tasks on ONN chips. It is anticipated that the OCDC can promote the novel accomplishment of ONNs in modern AI applications including autonomous driving, natural language processing, and scientific study.

Download Full-text