In the context of programming education, so-called auto-graders allow learners to receive automated feedback on their submissions. Because assessing learners' code typically involves executing the learners' untrusted code, this commonly used mechanism poses a significant security risk for these systems. Since auto-graders are mostly employed in the context of large-scale learning environments, such as universities or Massive Open Online Courses (MOOCs), security considerations are especially important. In this paper, we first introduce our auto-grader CodeOcean, which is regularly used in MOOCs with thousands of active learners, and in university contexts. As the execution of untrusted code can entail severe security implications, ensuring that the application contains no security vulnerabilities is essential. Hence, we partnered with a security consultancy to assess our auto-grader system landscape through a professional penetration test. This work presents the findings and countermeasures resulting from the performed security analysis for CodeOcean. We contextualize overarching enhancements for three main categories of threat vectors to auto-grader systems. Implementing these in any auto-grader system can improve the security and prevent learners from manipulating the assessment of their code. We also discuss the potential consequences of hardening an auto-grader, such as a reduced system performance. Therewith, we provide valuable recommendations for educators, researchers, and system designers to improve the security of auto-graders in the future, supporting their usage in even larger settings or in the context of exams.

Modern and future AI-based automotive applications, such as autonomous driving, require the efficient real-time processing of huge amounts of data from different sensors, like camera, radar, and LiDAR. In the ZuSE-KI-AVF project, multiple university, and industry partners collaborate to develop a novel massive parallel processor architecture, based on a cus-tomized RISC-V host processor, and an efficient high-performance vertical vector coprocessor. In addition, a software development framework is also provided to efficiently program AI-based sensor processing applications. The proposed processor system was verified and evaluated on a state-of-the-art UltraScale+ FPGA board, reaching a processing performance of up to 126.9 FPS, while executing the YOLO-LITE CNN on 224x224 input images. Further optimizations of the FPGA design and the realization of the processor system on a 22nm FDSOI CMOS technology are planned.

The digital transformation is challenging various areas of our everyday lives. Central aspects of our digital identities are our knowledge and experience, asserted by certificates, references and credentials. While digital credentials have primarily started to be used in the e-government and health sectors, acceptance is more and more transferring to the education sector. Within the area of digital credentials, a multitude of different projects and initiatives exist which are hard to follow and compare. Some researchers have attempted to perform systematic literature reviews. However, the scope of the review was often limited. Therefore, we create a conceptual framework to model the requirements of a digital education credential system by following both a conceptual-to-empirical approach and an empirical-to-conceptual approach. We perform an extensive literature review, focusing on identifying needs articulated in the relevant literature. We demonstrate the applicability and usability of the framework by developing a heatmap of requirements representing the current literature landscape. In a second step, we utilised the framework to interview subject matter experts and let them prioritise requirements for a digital education credential system. By comparing the heatmap results with the prioritisation of experts, we identified requirements, which are rarely found in the literature but rated as necessary by experts—making it most likely that they are typically overlooked when a new system is designed. We identified necessary, desirable and overlooked requirements and gave an indicator for prioritisation in our framework.

Architectures that first convert point clouds to a grid representation and then apply convolutional neural networks achieve good performance for radar-based object detection. However, the transfer from irregular point cloud data to a dense grid structure is often associated with a loss of information, due to the discretization and aggregation of points. In this paper, we propose a novel architecture, multi-scale KPPillarsBEV, that aims to mitigate the negative effects of grid rendering. Specifically, we propose a novel grid rendering method, KPBEV, which leverages the descriptive power of kernel point convolutions to improve the encoding of local point cloud contexts during grid rendering. In addition, we propose a general multi-scale grid rendering formulation to incorporate multi-scale feature maps into convolutional backbones of detection networks with arbitrary grid rendering methods. We perform extensive experiments on the nuScenes dataset and evaluate the methods in terms of detection performance and computational complexity. The proposed multi-scale KPPillarsBEV architecture outperforms the baseline by 5.37% and the previous state of the art by 2.88% in Car AP4.0 (average precision for a matching threshold of 4 meters) on the nuScenes validation set. Moreover, the proposed single-scale KPBEV grid rendering improves the Car AP4.0 by 2.90% over the baseline while maintaining the same inference speed.

Podcasts are a widely-used medium for communication and learning. One advantage of them is the possibility to pursue other activities while listening. Contrasting, Massive Open Online Courses (MOOCs) employ video-based teaching methods. Current research, however, challenges the interactivity and variation of teaching content in established MOOCs. This manuscript presents an experiment conducted with a podcast series deployed alongside a MOOC on cybersecurity. In our Static-Group Comparison, we identified a significant increase in learning success in weekly graded exercises (6.3%) and the course's final examination (6.4%) for learners exposing themselves to the podcast. Our first study results are promising in favor of multimedia learning. Hence, we present ideas for additional analysis and briefly outline which aspects of the results should be discussed in more depth.

This paper presents a method to learn the Cartesian velocity of objects using an object detection network on automotive radar data. The proposed method is self-supervised in terms of generating its own training signal for the velocities. Labels are only required for single-frame, oriented bounding boxes (OBBs). Labels for the Cartesian velocities or contiguous sequences, which are expensive to obtain, are not required. The general idea is to pre-train an object detection network without velocities using single-frame OBB labels, and then exploit the network’s OBB predictions on unlabelled data for velocity training. In detail, the network’s OBB predictions of the unlabelled frames are updated to the timestamp of a labelled frame using the predicted velocities and the distances between the updated OBBs of the unlabelled frame and the OBB predictions of the labelled frame are used to generate a self-supervised training signal for the velocities. The detection network architecture is extended by a module to account for the temporal relation of multiple scans and a module to represent the radars’ radial velocity measurements explicitly. A twostep approach of first training only OBB detection, followed by training OBB detection and velocities is used. Further, a pre-training with pseudo-labels generated from radar radial velocity measurements bootstraps the self-supervised method of this paper. Experiments on the publicly available nuScenes dataset show that the proposed method almost reaches the velocity estimation performance of a fully supervised training, but does not require expensive velocity labels. Furthermore, we outperform a baseline method which uses only radial velocity measurements as labels.

In this research, we formulate malware detection as a large-scale data-mining problem within Security Information and Event Management (SIEM) systems. We hypothesize that behavioral analysis of executable/process activities, such as file reads/writes, process creations, network connections, or registry modifications, enables the detection of advanced stealthy malware. To achieve this detection, we model processes behaviors as a set of directed acyclic graph streams and identify outliers in the set of graph streams. We enable this detection by conversion of the behavioral graph streams into documents, embedding using state-of-the-art Natural Language Processing model, and eventually performing novel outlier detection on the high dimensional vector representation of the documents. We evaluate our approach in a real-world setting, next to the SIEM system of a large-scale international enterprise (over 3TB of EDR logs). The proposed method has shown the capability to detect previously unknown threats.