اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدReducing DNN Labelling Cost using Surprise Adequacy: An Industrial Case Study for Autonomous Driving
278
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Deep Neural Networks (DNNs) are rapidly being adopted by the automotive industry, due to their impressive performance in tasks that are essential for autonomous driving. Object segmentation is one such task: its aim is to precisely locate boundaries of objects and classify the identified objects, helping autonomous cars to recognise the road environment and the traffic situation. Not only is this task safety critical, but developing a DNN based object segmentation module presents a set of challenges that are significantly different from traditional development of safety critical software. The development process in use consists of multiple iterations of data collection, labelling, training, and evaluation. Among these stages, training and evaluation are computation intensive while data collection and labelling are manual labour intensive. This paper shows how development of DNN based object segmentation can be improved by exploiting the correlation between Surprise Adequacy (SA) and model performance. The correlation allows us to predict model performance for inputs without manually labelling them. This, in turn, enables understanding of model performance, more guided data collection, and informed decisions about further training. In our industrial case study the technique allows cost savings of up to 50% with negligible evaluation inaccuracy. Furthermore, engineers can trade off cost savings versus the tolerable level of inaccuracy depending on different development phases and scenarios.
قيم البحث
اقرأ أيضاً
This paper employs correct-by-construction control synthesis, in particular controlled invariant set computations, for falsification. Our hypothesis is that if it is possible to compute a large enough controlled invariant set either for the actual sy
stem model or some simplification of the system model, interesting corner cases for other control designs can be generated by sampling initial conditions from the boundary of this controlled invariant set. Moreover, if falsifying trajectories for a given control design can be found through such sampling, then the controlled invariant set can be used as a supervisor to ensure safe operation of the control design under consideration. In addition to interesting initial conditions, which are mostly related to safety violations in transients, we use solutions from a dual game, a reachability game for the safety specification, to find falsifying inputs. We also propose optimization-based heuristics for input generation for cases when the state is outside the winning set of the dual game. To demonstrate the proposed ideas, we consider case studies from basic autonomous driving functionality, in particular, adaptive cruise control and lane keeping. We show how the proposed technique can be used to find interesting falsifying trajectories for classical control designs like proportional controllers, proportional integral controllers and model predictive controllers, as well as an open source real-world autonomous driving package.
There is a growing body of research on developing testing techniques for Deep Neural Networks (DNN). We distinguish two general modes of testing for DNNs: Offline testing where DNNs are tested as individual units based on test datasets obtained indep
endently from the DNNs under test, and online testing where DNNs are embedded into a specific application and tested in a close-loop mode in interaction with the application environment. In addition, we identify two sources for generating test datasets for DNNs: Datasets obtained from real-life and datasets generated by simulators. While offline testing can be used with datasets obtained from either sources, online testing is largely confined to using simulators since online testing within real-life applications can be time-consuming, expensive and dangerous. In this paper, we study the following two important questions aiming to compare test datasets and testing modes for DNNs: First, can we use simulator-generated data as a reliable substitute to real-world data for the purpose of DNN testing? Second, how do online and offline testing results differ and complement each other? Though these questions are generally relevant to all autonomous systems, we study them in the context of automated driving systems where, as study subjects, we use DNNs automating end-to-end control of cars steering actuators. Our results show that simulator-generated datasets are able to yield DNN prediction errors that are similar to those obtained by testing DNNs with real-life datasets. Further, offline testing is more optimistic than online testing as many safety violations identified by online testing could not be identified by offline testing, while large prediction errors generated by offline testing always led to severe safety violations detectable by online testing.
Mutation testing is used to evaluate the effectiveness of test suites. In recent years, a promising variation called extreme mutation testing emerged that is computationally less expensive. It identifies methods where their functionality can be entir
ely removed, and the test suite would not notice it, despite having coverage. These methods are called pseudo-tested. In this paper, we compare the execution and analysis times for traditional and extreme mutation testing and discuss what they mean in practice. We look at how extreme mutation testing impacts current software development practices and discuss open challenges that need to be addressed to foster industry adoption. For that, we conducted an industrial case study consisting of running traditional and extreme mutation testing in a large software project from the semiconductor industry that is covered by a test suite of more than 11,000 unit tests. In addition to that, we did a qualitative analysis of 25 pseudo-tested methods and interviewed two experienced developers to see how they write unit tests and gathered opinions on how useful the findings of extreme mutation testing are. Our results include execution times, scores, numbers of executed tests and mutators, reasons why methods are pseudo-tested, and an interview summary. We conclude that the shorter execution and analysis times are well noticeable in practice and show that extreme mutation testing supplements writing unit tests in conjunction with code coverage tools. We propose that pseudo-tested code should be highlighted in code coverage reports and that extreme mutation testing should be performed when writing unit tests rather than in a decoupled session. Future research should investigate how to perform extreme mutation testing while writing unit tests such that the results are available fast enough but still meaningful.
We investigate factors controlling DNN diversity in the context of the Google Cloud and YouTube-8M Video Understanding Challenge. While it is well-known that ensemble methods improve prediction performance, and that combining accurate but diverse pre
dictors helps, there is little knowledge on how to best promote & measure DNN diversity. We show that diversity can be cultivated by some unexpected means, such as model over-fitting or dropout variations. We also present details of our solution to the video understanding problem, which ranked #7 in the Kaggle competition (competing as the Yeti team).
Robust detection and tracking of objects is crucial for the deployment of autonomous vehicle technology. Image based benchmark datasets have driven development in computer vision tasks such as object detection, tracking and segmentation of agents in
the environment. Most autonomous vehicles, however, carry a combination of cameras and range sensors such as lidar and radar. As machine learning based methods for detection and tracking become more prevalent, there is a need to train and evaluate such methods on datasets containing range sensor data along with images. In this work we present nuTonomy scenes (nuScenes), the first dataset to carry the full autonomous vehicle sensor suite: 6 cameras, 5 radars and 1 lidar, all with full 360 degree field of view. nuScenes comprises 1000 scenes, each 20s long and fully annotated with 3D bounding boxes for 23 classes and 8 attributes. It has 7x as many annotations and 100x as many images as the pioneering KITTI dataset. We define novel 3D detection and tracking metrics. We also provide careful dataset analysis as well as baselines for lidar and image based detection and tracking. Data, development kit and more information are available online.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
