@phdthesis{26746, abstract = {{Previous research in proof-carrying hardware has established the feasibility and utility of the approach, and provided a concrete solution for employing it for the certification of functional equivalence checking against a specification, but fell short in connecting it to state-of-the-art formal verification insights, methods and tools. Due to the immense complexity of modern circuits, and verification challenges such as the state explosion problem for sequential circuits, this restriction of readily-available verification solutions severely limited the applicability of the approach in wider contexts. This thesis closes the gap between the PCH approach and current advances in formal hardware verification, provides methods and tools to express and certify a wide range of circuit properties, both functional and non-functional, and presents for the first time prototypes in which circuits that are implemented on actual reconfigurable hardware are verified with PCH methods. Using these results, designers can now apply PCH to establish trust in more complex circuits, by using more diverse properties which they can express using modern, efficient property specification techniques.}}, author = {{Wiersema, Tobias}}, keywords = {{Proof-Carrying Hardware, Formal Verification, Sequential Circuits, Non-Functional Properties, Functional Properties}}, pages = {{293}}, publisher = {{Paderborn University}}, title = {{{Guaranteeing Properties of Reconfigurable Hardware Circuits with Proof-Carrying Hardware}}}, year = {{2021}}, } @article{27045, abstract = {{Due to the lack of established real-world benchmark suites for static taint analyses of Android applications, evaluations of these analyses are often restricted and hard to compare. Even in evaluations that do use real-world apps, details about the ground truth in those apps are rarely documented, which makes it difficult to compare and reproduce the results. To push Android taint analysis research forward, this paper thus recommends criteria for constructing real-world benchmark suites for this specific domain, and presents TaintBench, the first real-world malware benchmark suite with documented taint flows. TaintBench benchmark apps include taint flows with complex structures, and addresses static challenges that are commonly agreed on by the community. Together with the TaintBench suite, we introduce the TaintBench framework, whose goal is to simplify real-world benchmarking of Android taint analyses. First, a usability test shows that the framework improves experts’ performance and perceived usability when documenting and inspecting taint flows. Second, experiments using TaintBench reveal new insights for the taint analysis tools Amandroid and FlowDroid: (i) They are less effective on real-world malware apps than on synthetic benchmark apps. (ii) Predefined lists of sources and sinks heavily impact the tools’ accuracy. (iii) Surprisingly, up-to-date versions of both tools are less accurate than their predecessors.}}, author = {{Luo, Linghui and Pauck, Felix and Piskachev, Goran and Benz, Manuel and Pashchenko, Ivan and Mory, Martin and Bodden, Eric and Hermann, Ben and Massacci, Fabio}}, issn = {{1382-3256}}, journal = {{Empirical Software Engineering}}, title = {{{TaintBench: Automatic real-world malware benchmarking of Android taint analyses}}}, doi = {{10.1007/s10664-021-10013-5}}, year = {{2021}}, } @article{21004, abstract = {{Automated machine learning (AutoML) supports the algorithmic construction and data-specific customization of machine learning pipelines, including the selection, combination, and parametrization of machine learning algorithms as main constituents. Generally speaking, AutoML approaches comprise two major components: a search space model and an optimizer for traversing the space. Recent approaches have shown impressive results in the realm of supervised learning, most notably (single-label) classification (SLC). Moreover, first attempts at extending these approaches towards multi-label classification (MLC) have been made. While the space of candidate pipelines is already huge in SLC, the complexity of the search space is raised to an even higher power in MLC. One may wonder, therefore, whether and to what extent optimizers established for SLC can scale to this increased complexity, and how they compare to each other. This paper makes the following contributions: First, we survey existing approaches to AutoML for MLC. Second, we augment these approaches with optimizers not previously tried for MLC. Third, we propose a benchmarking framework that supports a fair and systematic comparison. Fourth, we conduct an extensive experimental study, evaluating the methods on a suite of MLC problems. We find a grammar-based best-first search to compare favorably to other optimizers.}}, author = {{Wever, Marcel Dominik and Tornede, Alexander and Mohr, Felix and Hüllermeier, Eyke}}, issn = {{0162-8828}}, journal = {{IEEE Transactions on Pattern Analysis and Machine Intelligence}}, keywords = {{Automated Machine Learning, Multi Label Classification, Hierarchical Planning, Bayesian Optimization}}, pages = {{1--1}}, title = {{{AutoML for Multi-Label Classification: Overview and Empirical Evaluation}}}, doi = {{10.1109/tpami.2021.3051276}}, year = {{2021}}, } @article{21092, abstract = {{Automated Machine Learning (AutoML) seeks to automatically find so-called machine learning pipelines that maximize the prediction performance when being used to train a model on a given dataset. One of the main and yet open challenges in AutoML is an effective use of computational resources: An AutoML process involves the evaluation of many candidate pipelines, which are costly but often ineffective because they are canceled due to a timeout. In this paper, we present an approach to predict the runtime of two-step machine learning pipelines with up to one pre-processor, which can be used to anticipate whether or not a pipeline will time out. Separate runtime models are trained offline for each algorithm that may be used in a pipeline, and an overall prediction is derived from these models. We empirically show that the approach increases successful evaluations made by an AutoML tool while preserving or even improving on the previously best solutions.}}, author = {{Mohr, Felix and Wever, Marcel Dominik and Tornede, Alexander and Hüllermeier, Eyke}}, journal = {{IEEE Transactions on Pattern Analysis and Machine Intelligence}}, publisher = {{IEEE}}, title = {{{Predicting Machine Learning Pipeline Runtimes in the Context of Automated Machine Learning}}}, year = {{2021}}, } @inproceedings{21570, author = {{Tornede, Tanja and Tornede, Alexander and Wever, Marcel Dominik and Hüllermeier, Eyke}}, booktitle = {{Proceedings of the Genetic and Evolutionary Computation Conference}}, title = {{{Coevolution of Remaining Useful Lifetime Estimation Pipelines for Automated Predictive Maintenance}}}, year = {{2021}}, } @inbook{17905, abstract = {{This chapter concentrates on aspect-based sentiment analysis, a form of opinion mining where algorithms detect sentiments expressed about features of products, services, etc. We especially focus on novel approaches for aspect phrase extraction and classification trained on feature-rich datasets. Here, we present two new datasets, which we gathered from the linguistically rich domain of physician reviews, as other investigations have mainly concentrated on commercial reviews and social media reviews so far. To give readers a better understanding of the underlying datasets, we describe the annotation process and inter-annotator agreement in detail. In our research, we automatically assess implicit mentions or indications of specific aspects. To do this, we propose and utilize neural network models that perform the here-defined aspect phrase extraction and classification task, achieving F1-score values of about 80% and accuracy values of more than 90%. As we apply our models to a comparatively complex domain, we obtain promising results. }}, author = {{Kersting, Joschka and Geierhos, Michaela}}, booktitle = {{Natural Language Processing in Artificial Intelligence -- NLPinAI 2020}}, editor = {{Loukanova, Roussanka}}, pages = {{163----189 }}, publisher = {{Springer}}, title = {{{Towards Aspect Extraction and Classification for Opinion Mining with Deep Sequence Networks}}}, doi = {{10.1007/978-3-030-63787-3_6}}, volume = {{939}}, year = {{2021}}, } @misc{22304, author = {{Schott, Stefan}}, title = {{{Android App Analysis Benchmark Case Generation}}}, year = {{2021}}, } @inproceedings{22913, author = {{Hüllermeier, Eyke and Mohr, Felix and Tornede, Alexander and Wever, Marcel Dominik}}, location = {{Bilbao (Virtual)}}, title = {{{Automated Machine Learning, Bounded Rationality, and Rational Metareasoning}}}, year = {{2021}}, } @inproceedings{22927, author = {{Derrick, John and Doherty, Simon and Dongol, Brijesh and Schellhorn, Gerhard and Wehrheim, Heike}}, booktitle = {{Proceedings of the 35th International Symposium on Distributed Computing (DISC)}}, publisher = {{Schloß Dagstuhl}}, title = {{{On Strong Observational Refinement and Forward Simulation}}}, year = {{2021}}, } @inproceedings{22051, author = {{Kersting, Joschka and Geierhos, Michaela}}, booktitle = {{Proceedings of the 10th International Conference on Data Science, Technology and Applications (DATA 2021)}}, location = {{Online}}, pages = {{275----284}}, publisher = {{SCITEPRESS}}, title = {{{Well-being in Plastic Surgery: Deep Learning Reveals Patients' Evaluations}}}, year = {{2021}}, } @inproceedings{21953, author = {{Witschen, Linus Matthias and Wiersema, Tobias and Raeisi Nafchi, Masood and Bockhorn, Arne and Platzner, Marco}}, booktitle = {{Proceedings of International Symposium on Applied Reconfigurable Computing (ARC'21)}}, editor = {{Hannig, Frank and Derrien, Steven and Diniz, Pedro and Chillet, Daniel}}, location = {{Virtual conference}}, publisher = {{Springer Lecture Notes in Computer Science}}, title = {{{Timing Optimization for Virtual FPGA Configurations}}}, doi = {{10.1007/978-3-030-79025-7_4}}, year = {{2021}}, } @inproceedings{21598, abstract = {{Static analysis is used to automatically detect bugs and security breaches, and aids compileroptimization. Whole-program analysis (WPA) can yield high precision, however causes long analysistimes and thus does not match common software-development workflows, making it often impracticalto use for large, real-world applications.This paper thus presents the design and implementation ofModAlyzer, a novel static-analysisapproach that aims at accelerating whole-program analysis by making the analysis modular andcompositional. It shows how to computelossless, persisted summaries for callgraph, points-to anddata-flow information, and it reports under which circumstances this function-level compositionalanalysis outperforms WPA.We implementedModAlyzeras an extension to LLVM and PhASAR, and applied it to 12 real-world C and C++ applications. At analysis time,ModAlyzermodularly and losslessly summarizesthe analysis effect of the library code those applications share, hence avoiding its repeated re-analysis.The experimental results show that the reuse of these summaries can save, on average, 72% ofanalysis time over WPA. Moreover, because it is lossless, the module-wise analysis fully retainsprecision and recall. Surprisingly, as our results show, it sometimes even yields precision superior toWPA. The initial summary generation, on average, takes about 3.67 times as long as WPA.}}, author = {{Schubert, Philipp and Hermann, Ben and Bodden, Eric}}, booktitle = {{European Conference on Object-Oriented Programming (ECOOP)}}, title = {{{Lossless, Persisted Summarization of Static Callgraph, Points-To and Data-Flow Analysis}}}, year = {{2021}}, } @unpublished{30866, abstract = {{Automated machine learning (AutoML) strives for the automatic configuration of machine learning algorithms and their composition into an overall (software) solution - a machine learning pipeline - tailored to the learning task (dataset) at hand. Over the last decade, AutoML has developed into an independent research field with hundreds of contributions. While AutoML offers many prospects, it is also known to be quite resource-intensive, which is one of its major points of criticism. The primary cause for a high resource consumption is that many approaches rely on the (costly) evaluation of many machine learning pipelines while searching for good candidates. This problem is amplified in the context of research on AutoML methods, due to large scale experiments conducted with many datasets and approaches, each of them being run with several repetitions to rule out random effects. In the spirit of recent work on Green AI, this paper is written in an attempt to raise the awareness of AutoML researchers for the problem and to elaborate on possible remedies. To this end, we identify four categories of actions the community may take towards more sustainable research on AutoML, i.e. Green AutoML: design of AutoML systems, benchmarking, transparency and research incentives.}}, author = {{Tornede, Tanja and Tornede, Alexander and Hanselle, Jonas Manuel and Wever, Marcel Dominik and Mohr, Felix and Hüllermeier, Eyke}}, booktitle = {{arXiv:2111.05850}}, title = {{{Towards Green Automated Machine Learning: Status Quo and Future Directions}}}, year = {{2021}}, } @phdthesis{27284, author = {{Wever, Marcel Dominik}}, title = {{{Automated Machine Learning for Multi-Label Classification}}}, doi = {{10.17619/UNIPB/1-1302}}, year = {{2021}}, } @inbook{22052, abstract = {{In this study, we describe a text processing pipeline that transforms user-generated text into structured data. To do this, we train neural and transformer-based models for aspect-based sentiment analysis. As most research deals with explicit aspects from product or service data, we extract and classify implicit and explicit aspect phrases from German-language physician review texts. Patients often rate on the basis of perceived friendliness or competence. The vocabulary is difficult, the topic sensitive, and the data user-generated. The aspect phrases come with various wordings using insertions and are not noun-based, which makes the presented case equally relevant and reality-based. To find complex, indirect aspect phrases, up-to-date deep learning approaches must be combined with supervised training data. We describe three aspect phrase datasets, one of them new, as well as a newly annotated aspect polarity dataset. Alongside this, we build an algorithm to rate the aspect phrase importance. All in all, we train eight transformers on the new raw data domain, compare 54 neural aspect extraction models and, based on this, create eight aspect polarity models for our pipeline. These models are evaluated by using Precision, Recall, and F-Score measures. Finally, we evaluate our aspect phrase importance measure algorithm.}}, author = {{Kersting, Joschka and Geierhos, Michaela}}, booktitle = {{Natural Language Processing and Information Systems}}, editor = {{Kapetanios, Epaminondas and Horacek, Helmut and Métais, Elisabeth and Meziane, Farid}}, location = {{Saarbrücken, Germany}}, pages = {{231----242}}, publisher = {{Springer}}, title = {{{Human Language Comprehension in Aspect Phrase Extraction with Importance Weighting}}}, volume = {{12801}}, year = {{2021}}, } @inproceedings{28199, author = {{Pauck, Felix and Wehrheim, Heike}}, booktitle = {{2021 IEEE 21st International Working Conference on Source Code Analysis and Manipulation (SCAM)}}, title = {{{Jicer: Simplifying Cooperative Android App Analysis Tasks}}}, doi = {{10.1109/scam52516.2021.00031}}, year = {{2021}}, } @inproceedings{23709, author = {{Chen, Wei-Fan and Al Khatib, Khalid and Stein, Benno and Wachsmuth, Henning}}, booktitle = {{Findings of the Association for Computational Linguistics: EMNLP 2021}}, pages = {{2683 -- 2693}}, title = {{{Controlled Neural Sentence-Level Reframing of News Articles}}}, year = {{2021}}, } @inproceedings{22229, author = {{Alshomary, Milad and Syed, Shahbaz and Potthast, Martin and Wachsmuth, Henning}}, booktitle = {{Proceedings of the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021)}}, location = {{Online}}, pages = {{1816–1827}}, publisher = {{Association for Computational Linguistics}}, title = {{{Argument Undermining: Counter-Argument Generation by Attacking Weak Premises}}}, doi = {{10.18653/v1/2021.findings-acl.159}}, year = {{2021}}, } @inproceedings{21198, author = {{Hanselle, Jonas Manuel and Tornede, Alexander and Wever, Marcel Dominik and Hüllermeier, Eyke}}, location = {{Delhi, India}}, title = {{{Algorithm Selection as Superset Learning: Constructing Algorithm Selectors from Imprecise Performance Data}}}, year = {{2021}}, } @article{27841, abstract = {{Verification of software and processor hardware usually proceeds separately, software analysis relying on the correctness of processors executing machine instructions. This assumption is valid as long as the software runs on standard CPUs that have been extensively validated and are in wide use. However, for processors exploiting custom instruction set extensions to meet performance and energy constraints the validation might be less extensive, challenging the correctness assumption. In this paper we present a novel formal approach for hardware/software co-verification targeting processors with custom instruction set extensions. We detail two different approaches for checking whether the hardware fulfills the requirements expected by the software analysis. The approaches are designed to explore a trade-off between generality of the verification and computational effort. Then, we describe the integration of software and hardware analyses for both techniques and describe a fully automated tool chain implementing the approaches. Finally, we demonstrate and compare the two approaches on example source code with custom instructions, using state-of-the-art software analysis and hardware verification techniques.}}, author = {{Jakobs, Marie-Christine and Pauck, Felix and Platzner, Marco and Wehrheim, Heike and Wiersema, Tobias}}, journal = {{IEEE Access}}, keywords = {{Software Analysis, Abstract Interpretation, Custom Instruction, Hardware Verification}}, publisher = {{IEEE}}, title = {{{Software/Hardware Co-Verification for Custom Instruction Set Processors}}}, doi = {{10.1109/ACCESS.2021.3131213}}, year = {{2021}}, }