Georg Moser

Albert Nössig, Tobias Hell, Georg Moser

In this paper, we present a modular methodology that combines state-of-the-art methods in (stochastic) machine learning with traditional methods in rule learning to provide efficient and scalable algorithms for the classification of vast data sets, while remaining explainable. Apart from evaluating our approach on the common large scale data sets MNIST, Fashion-MNIST and IMDB, we present novel results on explainable classifications of dental bills. The latter case study stems from an industrial collaboration with Allianz Private Krankenversicherungs-Aktiengesellschaft which is an insurance company offering diverse services in Germany.

Armin Walch, Georg Moser, Berry Schoenmakers et al.

We study the fully automated amortised analysis of purely functional data structures like skew heaps, as well as weight- and rank-biased leftist heaps. For that we generalise earlier works on automated amortised resource analysis by developing a type inference based approach with a generic type system. This allows for modular reasoning and the inference of precise and optimal cost bounds. More specifically, we extend the work on the ATLAS system by Leutgeb et al. which was developed to cover the analysis of splay trees and some closely related data structures. To enable the analysis of skew heaps, however, and the even more challenging (amortised) analysis of leftist heaps, we have developed a range of new techniques for type-based automated analysis. By introducing a generic type system we allow for arbitrary (classes of) potential functions, compared to the use of hard-coded potential functions in ATLAS, which we have implemented in Haskell in an entirely modular way. We have also greatly enhanced the existing type inference algorithm by extensions in multiple directions, including path-sensitive reasoning, data structure invariants, and template parameters for piecewise defined potential functions. We show how our newly developed system supports the use of all known potential functions for analysing skew heaps and leftist heaps, confirming the known bounds.

Samuel Frontull, Georg Moser

This paper explores the impact of different back-translation approaches on machine translation for Ladin, specifically the Val Badia variant. Given the limited amount of parallel data available for this language (only 18k Ladin-Italian sentence pairs), we investigate the performance of a multilingual neural machine translation model fine-tuned for Ladin-Italian. In addition to the available authentic data, we synthesise further translations by using three different models: a fine-tuned neural model, a rule-based system developed specifically for this language pair, and a large language model. Our experiments show that all approaches achieve comparable translation quality in this low-resource scenario, yet round-trip translations highlight differences in model performance.

Albert Nössig, Tobias Hell, Georg Moser

State-of-the-art results in typical classification tasks are mostly achieved by unexplainable machine learning methods, like deep neural networks, for instance. Contrarily, in this paper, we investigate the application of rule learning methods in such a context. Thus, classifications become based on comprehensible (first-order) rules, explaining the predictions made. In general, however, rule-based classifications are less accurate than state-of-the-art results (often significantly). As main contribution, we introduce a voting approach combining both worlds, aiming to achieve comparable results as (unexplainable) state-of-the-art methods, while still providing explanations in the form of deterministic rules. Considering a variety of benchmark data sets including a use case of significant interest to insurance industries, we prove that our approach not only clearly outperforms ordinary rule learning methods, but also yields results on a par with state-of-the-art outcomes.

Georg Moser, Michael Schaper

In recent years, quantum computing has gained a substantial amount of momentum, and the capabilities of quantum devices are continually expanding and improving. Nevertheless, writing a quantum program from scratch remains tedious and error-prone work, showcasing the clear demand for automated tool support. We present Qet, a fully automated static program analysis tool that yields a precise expected cost analysis of mixed classical-quantum programs. Qet supports programs with advanced features like mid-circuit measurements and classical control flow. The methodology of our prototype implementation is based on a recently proposed quantum expectation transformer framework, generalising Dijkstra's predicate transformer and Hoare logic. The prototype implementation Qet is evaluated on a number of case studies taken from the literature and online references. Qet is able to fully automatically infer precise upper bounds on the expected costs that previously could only be derived by tedious manual calculations.

Albert Nössig, Tobias Hell, Georg Moser

In this paper, we present an innovative iterative approach to rule learning specifically designed for (but not limited to) text-based data. Our method focuses on progressively expanding the vocabulary utilized in each iteration resulting in a significant reduction of memory consumption. Moreover, we introduce a Value of Confidence as an indicator of the reliability of the generated rules. By leveraging the Value of Confidence, our approach ensures that only the most robust and trustworthy rules are retained, thereby improving the overall quality of the rule learning process. We demonstrate the effectiveness of our method through extensive experiments on various textual as well as non-textual datasets including a use case of significant interest to insurance industries, showcasing its potential for real-world applications.

Guillaume Bonfante, Georg Moser

The DICE workshop explores the area of Implicit Computational Complexity (ICC), which grew out from several proposals to use logic and formal methods to provide languages for complexity-bounded computation (e.g. Ptime, Logspace computation). It aims at studying the computational complexity of programs without referring to external measuring conditions or a particular machine model, but only by considering language restrictions or logical/computational principles entailing complexity properties. The FOPARA workshop serves as a forum for presenting original research results that are relevant to the analysis of resource (e.g. time, space, energy) consumption by computer programs. The workshop aims to bring together the researchers that work on foundational issues with the researchers that focus more on practical results. Therefore, both theoretical and practical contributions are encouraged. We also encourage papers that combine theory and practice. Given the complementarity and the synergy between these two communities, and following the successful experience of co-location of DICE-FOPARA 2015 in London at ETAPS 2015, we hold these two workshops together at ETAPS 2017, which takes place in Uppsala, Sweden. The provided proceedings collect the papers accepted at the workshop.