Systematically testing models learned from neural networks remains a crucial unsolved barrier to successfully justify safety for autonomous vehicles engineered using data-driven approach. We propose quantitative k-projection coverage as a metric to mediate combinatorial explosion while guiding the data sampling process. By assuming that domain experts propose largely independent environment conditions and by associating elements in each condition with weights, the product of these conditions forms scenarios, and one may interpret weights associated with each equivalence class as relative importance. Achieving full k-projection coverage requires that the data set, when being projected to the hyperplane formed by arbitrarily selected k-conditions, covers each class with number of data points no less than the associated weight. For the general case where scenario composition is constrained by rules, precisely computing k-projection coverage remains in NP. In terms of finding minimum test cases to achieve full coverage, we present theoretical complexity for important sub-cases and an encoding to 0-1 integer programming. We have implemented a research prototype that generates test cases for a visual object detection unit in automated driving, demonstrating the technological feasibility of our proposed coverage criterion.
IBM CPLEX OptimizationStudio: https
We study the problem of formal verification of Binarized Neural Networks (BNN), which have recently been proposed as a energyefficient alternative to traditional learning networks. The verification of BNNs, using the reduction to hardware verification, can be even more scalable by factoring computations among neurons within the same layer. By proving the NP-hardness of finding optimal factoring as well as the hardness of PTAS approximability, we design polynomial-time search heuristics to generate factoring solutions. The overall framework allows applying verification techniques to moderately-sized BNNs for embedded devices with thousands of neurons and inputs.
This paper proposes Kalman-filter drift removal (DR) and Heron-bilateration location estimation (LE) to significantly reduce the received signal strength index (RSSI) drift, localization error, computational complexity, and deployment cost of conventional radio frequency identification (RFID) indoor positioning systems without any sacrifice of localization granularity and accuracy. By means of only one portable RFID reader as the targeted device and only one pair of active RFID tags as the border-deployed landmarks, this paper develops a real-time portable RFID indoor positioning device and costeffective scalable RFID indoor positioning infrastructure, based on Kalman-filter DR, Heron-bilateration LE, and four novel preprocessing/postprocessing techniques. Experimental results reveal that the proposed Kalman-filter DR method is faster and better to converge the distance measurement (DM) error than conventional probability/statistics in terms of various relative distances under certain RSSI drift effect condition, and the proposed Heron-bilateration LE method is also faster and better to converge the LE error than conventional proximity pattern matching and trilateration in terms of three or more landmarks under certain DM error condition. On the other hand, a portable RFID indoor positioning device is smoothly implemented on an Android smartphone platform attached with a portable Bluetooth-based RFID reader.Index Terms-Drift removal (DR), indoor positioning, location estimation (LE), radio frequency identification (RFID).
Artificial neural networks (NN) are instrumental in realizing highly-automated driving functionality. An overarching challenge is to identify best safety engineering practices for NN and other learning-enabled components. In particular, there is an urgent need for an adequate set of metrics for measuring allimportant NN dependability attributes. We address this challenge by proposing a number of NN-specific and efficiently computable metrics for measuring NN dependability attributes including robustness, interpretability, completeness, and correctness.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.