34711 research outputs found
Sort by
Towards optimized AI strategies for cryptocurrency trading
Get PDFThis thesis investigates the application of advanced AI techniques, specifically Double Deep Q-Networks, for developing optimized cryptocurrency trading strategies. We explore the efficacy of two distinct neural network architectures as Q-function approximators: a novel time-series model, TimesNet, and a conventional 1D CNN. The study employs historical price data (OHLCV) for Bitcoin and Ethereum, evaluating performance with both a basic feature set and an extended set incorporating technical indicators and blockchain-specific metrics. Our experiments demonstrate that DDQN agents, particularly those leveraging the TimesNet architecture, can learn complex trading behaviors and achieve significant outperformance compared to traditional Buy & Hold benchmarks and CNN-based agents across diverse market conditions. The research highlights the potential of sophisticated temporal modeling and reinforcement learning in navigating the volatile cryptocurrency landscape, while also underscoring the nuanced impact of feature engineering on strategy performance across different market regimes and assets
Future use of carbon management to achieve emissions reductions and removals in Switzerland : scenario and modeling analysis
No full textLiving with pulmonary fibrosis : how affected people experience disease-related information, health services and self-management strategies
Get PDFIntroduction: People with pulmonary fibrosis (PF) experience gaps in care, educational resources and self-management strategies throughout their journey. We sought to identify gaps in care and information, determine essential information, examine sources of information and assess preferred modes of delivery for patient education and self-management (PESM) resources and interventions.
Methods: In this qualitative study, we conducted interviews with people with PF using predefined, literature-based categories (“staying well with PF”, “keeping fit & strong with PF”, “using oxygen therapy”, symptom management “breathlessness”, “cough”, “fatigue”, “anxiety, depression and panic”) and categories derived from own experience (“sources of information”, “preferred modes of education delivery”). Interviews were analysed using deductive-inductive content analysis.
Results: We interviewed 11 individuals with PF (one woman), median age of 73 years (range 52–80) and time since diagnosis ranging from 1 to 10 years. Interviews lasted 40–70 min. A priori-defined saturation was reached for each predefined topic. Patients with PF relied on healthcare professionals (HCPs) as their primary information source. However, for sensitive topics such as life expectancy, death or when time with HCPs was limited, they often turned to internet searches with little success. There was a need for guidance on managing antifibrotic side effects, including adaptation of nutrition. While exercise was valued, it lacked structure, particularly at home. Self-management strategies for cough and fatigue remained insufficient. Digital resources were considered beneficial but maintaining in-person interactions with HCPs is essential.
Conclusion: This in-depth analysis highlights how people with PF comprehend patient education content and which taught or self-acquired self-management techniques they implement in their lives, including beliefs and concerns. These insights are essential for refining current and future PESM programmes and support the creation of blended learning and digital resources, with HCPs offering guidance on their use, to further support people with PF in self-management
The regress of uncertainty and the forecasting paradox
Get PDFWe show that epistemic uncertainty–our iterated ignorance about our own ignorance–inevitably thickens statistical tails, even under perceived thin-tailed environments from past realizations. Any claim of precise risk carries a margin of error, and that margin itself is uncertain, in an infinite regress of doubt. This “errors-on-errors” mechanism rules out thin-tailed certainty: predictive laws must be heavier-tailed than their in-sample counterparts. The result is the Forecasting Paradox: the future is structurally more extreme than the past. This insight collapses branching scenarios into a single heavy-tailed forecast, with direct implications for risk management, scientific modeling, and AI safety
Gerechtfertigte Entlassung wegen Absenzen
No full textVersäumt eine erkrankte Arbeitnehmerin ihre Pflicht, die Arbeitgeberin kontinuierlich über ihre Arbeitsunfähigkeit zu unterrichten, riskiert sie eine fristlose Entlassung
Exploratory analysis of direct load control policies for heat pumps in the future Swiss electricity system
No full textNew flexible electrical loads such as heat pumps present major demand-side flexibility opportunities thanks to their load shifting potential. In this study, we analyze to what extent the direct load control (DLC) of heat pumps within realistic bounds facilitates their integration in the Swiss energy system as it decarbonizes, by coupling a bottom-up building stock model and a high-resolution electricity system model. We rely on recently offered DLC contracts by Swiss distribution system operators to set these bounds and study the impact of different DLC settings. We find that contract parameters such as allowing preheating and prolonging shifting windows significantly impact how much DLC is performed. However, even high levels of DLC lead to small impacts on the electricity system's cost and renewables' integration in Switzerland, possibly only having a significant effect on peak electricity prices
The influence of difficult child temperament and parenting stress on parental problematic smartphone use in early childhood : a moderated mediation analysis
No full textSmartphone use has become an integral part of daily life and is also common among parents of young children. As parental problematic smartphone use can impact parents’ and children’s well-being, it is important to examine the determinants of this issue. This study examined the association between difficult temperament in toddlers, parenting stress, and parental problematic smartphone use. Results of a moderated mediation analysis utilizing data from 261 mothers and fathers of 14-month-old children showed that parenting stress fully mediated the association between difficult child temperament and parental problematic smartphone use. Parental gender did not moderate the association between parenting stress and problematic smartphone use. Parents who rated their child’s temperament as more difficult reported more parenting stress, which in turn led to higher levels of problematic smartphone use. Parenting stress is thus an important determinant of parental problematic smartphone use. Further research on stress management and parental smartphone use is needed to develop targeted interventions that promote the well-being of parents and children. Parental smartphone use should be considered holistically by taking various influencing factors and their interplay into account rather than examining them in isolation
Creative reflection spaces in human-machine collaboration : cultivating non-cognitive competences in BANI times
No full textIn an era defined by rapid technological acceleration and increasing integration of artificial intelligence (AI) in linguistic and communicative domains, the question of how human creativity and reflective capacities intersect with AI becomes central. This contribution explores the role of creative reflection as a methodological and epistemological tool for cultivating non-cognitive competences in applied linguistics contexts. Drawing on interdisciplinary frameworks integrating non-cognitive competences (IDG, 2025; Sunder et al., 2024; OECD, 2019; Gutman et al., 2013; Kusuma et al., 2023), the presentation synthesizes insights from Linda Candy's Creative Reflective Practice (2020)—who further developed the approach of Donald Schön's Reflective Practice (1987/2016)—the multi-perspective view of Creative Agility (ZHAW, 2024/2025), and the transcendent potential of spiritual poetry (Borghoff, 2021; Leopardi et al., 2013; Friedman, 2016; Marcato, 2023; Walker, 2013/2019/2021; Whickman, 2016; Robinson & Cussen, 2022; Spies, 2012; Jastrzñebski, 2023; brückenwege, 2025).
First, the study examines how creative approaches to reflection provide applied linguists, translators, educators, and creative content creators with practices, interventions and tools to critically assess their interactions with AI systems. Schön's concept of reflection-in-action and reflection-on-action is revisited in light of creative reflective practice according to Candy (2020), human-AI collaboration, emphasizing the dynamic interplay between intuition, problem-framing, and real-time decision-making in AI-mediated workflows. Second, Creative Agility is positioned as a competence framework that enables practitioners to navigate the brittle, anxious, nonlinear and incomprehensible environments created by AI integration in language education, translation technologies, and content creation. Creagile reflection fosters adaptive thinking and empowers linguists to maintain authentic, human-centric communication amidst algorithmic standardization. Third, spiritually inspired poetry is introduced as an innovative lens to explore the often-overlooked transcendent dimensions of reflective creativity. In applied linguistics, spiritual poetic reflection can serve as a meta-cognitive tool that transcends logical reasoning, enabling practitioners to articulate tacit knowledge, emotional depth, and ethical intuition in their engagement with AI systems. The study aligns these creative reflection approaches with the non-cognitive categories of the Inner Development Goals Framework, this is being, relating, collaborating and acting (IDG 2025)—particularly focusing on integrity and authenticity (intra-personal competences), empathy and compassion, communication (inter-personal competences), and creativity (implementation competences) (Neuhold & Borghoff, forthcoming 2025). The findings argue that creative reflection is not only a methodological tool but also a holistic, transformative practice that creatively bridges the human and the technological, fostering resilient and ethically grounded innovating linguistic ecosystems (for ecosystems see e.g. Theodoraki et al., 2022; Theodoraki, 2018; Theodoraki & Messeghem, 2017; Tabas et al., 2025).
The presentation concludes with an exploration of how creative reflective practices can be implemented in applied linguistics education, AI-enhanced translation workflows, and content creation strategies. By fostering non-cognitive competences, reflective spaces emerge as vital arenas for preserving human agency, ethical responsibility, and emotional intelligence in AI-mediated linguistic landscapes. This contribution is positioned at the intersection of creativity, technology, and creative reflective praxis, offering actionable insights for linguists, translators, educators, and researchers navigating the complexities of human-machine collaboration
The neurobench framework for benchmarking neuromorphic computing algorithms and systems
No full textNeuromorphic computing shows promise for advancing computing efficiency and capabilities of AI applications using brain-inspired principles. However, the neuromorphic research field currently lacks standardized benchmarks, making it difficult to accurately measure technological advancements, compare performance with conventional methods, and identify promising future research directions. This article presents NeuroBench, a benchmark framework for neuromorphic algorithms and systems, which is collaboratively designed from an open community of researchers across industry and academia. NeuroBench introduces a common set of tools and systematic methodology for inclusive benchmark measurement, delivering an objective reference framework for quantifying neuromorphic approaches in both hardware-independent and hardware-dependent settings. For latest project updates, visit the project website (neurobench.ai)
Extremismusstudie für die Stadt St.Gallen
No full textZur Beantwortung des vorliegenden Postulats des St. Galler Stadtparlaments und zur Einschätzung des Umgangs mit Extremismusvorfällen wurde der vorliegende Bericht in Auftrag gegeben. Da der Gemeinderat der Stadt Zürich zur selben Zeit ähnliche Postulate zur Extremismusthematik überwiesen hatte, schlossen sich die Auftraggebenden aus der Stadt St. Gallen und Zürich für den Auftrag des Extremismusberichts zusammen, was Synergien für die übergreifenden Fragestellungen brachte. Der Extremismusbericht des ZHAW-Teams wurde von einer Begleitgruppe mit Fachpersonen aus den beiden Städten koordiniert, unterstützt und begleitet. Am Schluss entschied sich die Begleitgruppe für zwei getrennte Berichte für die Stadt Zürich und die Stadt St. Gallen. In beiden Städten wurden extremistische Vorfälle analysiert, die Interventionen der Behörden untersucht und Empfehlungen abgeleitet, um präventive Handlungsmöglichkeiten gegen gewaltbereiten Extremismus aufzuzeigen.
Der Extremismusbericht soll die Vorfälle in die aktuellen Entwicklungen und Extremismustrends einordnen, Fragen zu Radikalisierungsprozessen und extremistischen Organisationen klären sowie Einblicke in deren Kommunikation mittels Social Media-Analyse geben. Ein weiterer Fokus liegt auf der Bewertung der Reaktionen und Interventionen der städtischen Behörden zur Bestimmung des Optimierungs- und Weiterbildungsbedarfs.
Die Ergebnisse zeigen, wie stark sich unterschiedliche Extremismusformen in aktuellen Themenfeldern wie etwa der Queerfeindlichkeit oder dem Gaza-Konflikt im physischen sowie digitalen Raum überlagern. Gleichzeitig haben diese Themenfelder auch eine Stellvertreterfunktion für grössere gesellschaftliche Debatten rund um Minderheitenrechte oder Fragen von Diskriminierung und geopolitischen Verstrickungen in Konflikte und führen zu gesellschaftlichen Polarisierungen. Im Weiteren beschäftigen Staatsverweigerer:innen Behörden auf verschiedenen Ebenen der Verwaltung.
Aufgrund der bearbeiteten Fragestellungen zeichnet sich für die Stadt St. Gallen eine wenig ausgeprägte Bedarfslage bzgl. Extremismuspotentialen und Radikalisierungsprozessen im Jugendbereich ab, da es kaum Hinweise auf Radikalisierung von Minderjährigen gibt. Das bestehende Beratungs- und Unterstützungsangebot wird als ausreichend beurteilt, doch empfiehlt sich eine verbesserte Koordination mit der kantonalen Fachstelle FAREX sowie der Einbezug respektive Berücksichtigung der Staatsverweigerer:innen-Thematik in deren Beratungsangebot