NPC: <u>N</u> euron <u>P</u> ath <u>C</u> overage via Characterizing Decision Logic of Deep Neural Networks

نویسندگان

چکیده

Deep learning has recently been widely applied to many applications across different domains, e.g., image classification and audio recognition. However, the quality of Neural Networks (DNNs) still raises concerns in practical operational environment, which calls for systematic testing, especially safety-critical scenarios. Inspired by software a number structural coverage criteria are designed proposed measure test adequacy DNNs. due blackbox nature DNN, existing difficult interpret, making it hard understand underlying principles these criteria. The relationship between decision logic DNNs is unknown. Moreover, recent studies have further revealed non-existence correlation DNN defect detection, posts on what suitable testing criterion should be. In this article, we propose interpretable through constructing structure DNN. Mirroring control flow graph traditional program, first extract from based its interpretation, where path represents Based data graph, two variants cases exercising logic. higher coverage, more diverse expected be explored. Our large-scale evaluation results demonstrate that: effective characterizing also sensitive with errors, including natural errors adversarial examples, strongly correlate output impartiality.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Using Artificial Neural Networks in NPC Decision- Making Process

The present paper presents the use of ANN (Artifical Neural Networks) in NPC (Non-Player Character) decisionmaking process in RPG (Role-Playing Games) electronic games. It mainly focus on the making of decision of attacking or not a certain opponent by analyzing the NPC and its opponent internal state. Keywordsgames; artifical neural networks; decision-making process; NPC

متن کامل

Harnessing Deep Neural Networks with Logic Rules

Combining deep neural networks with structured logic rules is desirable to harness flexibility and reduce unpredictability of the neural models. We propose a general framework capable of enhancing various types of neural networks (e.g., CNNs and RNNs) with declarative first-order logic rules. Specifically, we develop an iterative distillation method that transfers the structured information of ...

متن کامل

Choice by Elimination via Deep Neural Networks

We introduce Neural Choice by Elimination, a new framework that integrates deep neural networks into probabilistic sequential choice models for learning to rank. Given a set of items to chose from, the elimination strategy starts with the whole item set and iteratively eliminates the least worthy item in the remaining subset. We prove that the choice by elimination is equivalent to marginalizin...

متن کامل

Nonlinear Interference Mitigation via Deep Neural Networks

A neural-network-based approach is presented to efficiently implement digital backpropagation (DBP). For a 32×100 km fiber-optic link, the resulting “learned” DBP significantly reduces the complexity compared to conventional DBP implementations. OCIS codes: (060.0060) Fiber optics and optical communications, (060.2330) Fiber optics communications.

متن کامل

Methods of Characterizing Neural Networks

The UCLA biophysics neuroimaging laboratory was created to develop new methods to approach systems neuroscience. Currently, the lab is endeavoring to understand how neural networks develop and process information. The organism of focus is the zebrafish, and methods of imaging are confocal and two-photon imaging. The primary goal is to use calcium imaging to create a three dimensional image of t...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: ACM Transactions on Software Engineering and Methodology

سال: 2022

ISSN: ['1049-331X', '1557-7392']

DOI: https://doi.org/10.1145/3490489