Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userTowards Characterizing Adversarial Defects of Deep Learning Software from the Lens of Uncertainty
115
0
0.0
(
0
)
Added by
Xiyue Zhang
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Over the past decade, deep learning (DL) has been successfully applied to many industrial domain-specific tasks. However, the current state-of-the-art DL software still suffers from quality issues, which raises great concern especially in the context of safety- and security-critical scenarios. Adversarial examples (AEs) represent a typical and important type of defects needed to be urgently addressed, on which a DL software makes incorrect decisions. Such defects occur through either intentional attack or physical-world noise perceived by input sensors, potentially hindering further industry deployment. The intrinsic uncertainty nature of deep learning decisions can be a fundamental reason for its incorrect behavior. Although some testing, adversarial attack and defense techniques have been recently proposed, it still lacks a systematic study to uncover the relationship between AEs and DL uncertainty. In this paper, we conduct a large-scale study towards bridging this gap. We first investigate the capability of multiple uncertainty metrics in differentiating benign examples (BEs) and AEs, which enables to characterize the uncertainty patterns of input data. Then, we identify and categorize the uncertainty patterns of BEs and AEs, and find that while BEs and AEs generated by existing methods do follow common uncertainty patterns, some other uncertainty patterns are largely missed. Based on this, we propose an automated testing technique to generate multiple types of uncommon AEs and BEs that are largely missed by existing techniques. Our further evaluation reveals that the uncommon data generated by our method is hard to be defended by the existing defense techniques with the average defense success rate reduced by 35%. Our results call for attention and necessity to generate more diverse data for evaluating quality assurance solutions of DL software.
rate research
Read More
Deep learning (DL) techniques have gained significant popularity among software engineering (SE) researchers in recent years. This is because they can often solve many SE challenges without enormous manual feature engineering effort and complex domain knowledge. Although many DL studies have reported substantial advantages over other state-of-the-art models on effectiveness, they often ignore two factors: (1) replicability - whether the reported experimental result can be approximately reproduced in high probability with the same DL model and the same data; and (2) reproducibility - whether one reported experimental findings can be reproduced by new experiments with the same experimental protocol and DL model, but different sampled real-world data. Unlike traditional machine learning (ML) models, DL studies commonly overlook these two factors and declare them as minor threats or leave them for future work. This is mainly due to high model complexity with many manually set parameters and the time-consuming optimization process. In this study, we conducted a literature review on 93 DL studies recently published in twenty SE journals or conferences. Our statistics show the urgency of investigating these two factors in SE. Moreover, we re-ran four representative DL models in SE. Experimental results show the importance of replicability and reproducibility, where the reported performance of a DL model could not be replicated for an unstable optimization process. Reproducibility could be substantially compromised if the model training is not convergent, or if performance is sensitive to the size of vocabulary and testing data. It is therefore urgent for the SE community to provide a long-lasting link to a replication package, enhance DL-based solution stability and convergence, and avoid performance sensitivity on different sampled data.
Given the current transformative potential of research that sits at the intersection of Deep Learning (DL) and Software Engineering (SE), an NSF-sponsored community workshop was conducted in co-location with the 34th IEEE/ACM International Conference on Automated Software Engineering (ASE19) in San Diego, California. The goal of this workshop was to outline high priority areas for cross-cutting research. While a multitude of exciting directions for future work were identified, this report provides a general summary of the research areas representing the areas of highest priority which were discussed at the workshop. The intent of this report is to serve as a potential roadmap to guide future work that sits at the intersection of SE & DL.
The development of scientific software is, more than ever, critical to the practice of science, and this is accompanied by a trend towards more open and collaborative efforts. Unfortunately, there has been little investigation into who is driving the evolution of such scientific software or how the collaboration happens. In this paper, we address this problem. We present an extensive analysis of seven open-source scientific software projects in order to develop an empirically-informed model of the development process. This analysis was complemented by a survey of 72 scientific software developers. In the majority of the projects, we found senior research staff (e.g. professors) to be responsible for half or more of commits (an average commit share of 72%) and heavily involved in architectural concerns (seniors were more likely to interact with files related to the build system, project meta-data, and developer documentation). Juniors (e.g.graduate students) also contribute substantially -- in one studied project, juniors made almost 100% of its commits. Still, graduate students had the longest contribution periods among juniors (with 1.72 years of commit activity compared to 0.98 years for postdocs and 4 months for undergraduates). Moreover, we also found that third-party contributors are scarce, contributing for just one day for the project. The results from this study aim to help scientists to better understand their own projects, communities, and the contributors behavior, while paving the road for future software engineering research
In 2006, Geoffrey Hinton proposed the concept of training Deep Neural Networks (DNNs) and an improved model training method to break the bottleneck of neural network development. More recently, the introduction of AlphaGo in 2016 demonstrated the powerful learning ability of deep learning and its enormous potential. Deep learning has been increasingly used to develop state-of-the-art software engineering (SE) research tools due to its ability to boost performance for various SE tasks. There are many factors, e.g., deep learning model selection, internal structure differences, and model optimization techniques, that may have an impact on the performance of DNNs applied in SE. Few works to date focus on summarizing, classifying, and analyzing the application of deep learning techniques in SE. To fill this gap, we performed a survey to analyse the relevant studies published since 2006. We first provide an example to illustrate how deep learning techniques are used in SE. We then summarize and classify different deep learning techniques used in SE. We analyzed key optimization technologies used in these deep learning models, and finally describe a range of key research topics using DNNs in SE. Based on our findings, we present a set of current challenges remaining to be investigated and outline a proposed research road map highlighting key opportunities for future work.
In this paper we introduce the notion of Modal Software Engineering: automatically turning sequential, deterministic programs into semantically equivalent programs efficiently operating on inputs coming from multiple overlapping worlds. We are drawing an analogy between modal logics, and software application domains where multiple sets of inputs (multiple worlds) need to be processed efficiently. Typically those sets highly overlap, so processing them independently would involve a lot of redundancy, resulting in lower performance, and in many cases intractability. Three application domains are presented: reasoning about feature-based variability of Software Product Lines (SPLs), probabilistic programming, and approximate programming.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
