The Artificial and Engineering Intelligence Lab (AEIL) is focused on pioneering research at the intersection of artificial intelligence and engineering disciplines. The lab aims to enhance engineering processes by integrating advanced AI techniques, improving system efficiency and innovation. Researchers at AEIL develop new methodologies to leverage AI’s computational strengths, applying them to complex engineering challenges such as automation, optimization, and adaptive control. The lab’s projects range from smart infrastructure systems to intelligent manufacturing processes, exploring practical applications that drive technological advancement. By emphasizing interdisciplinary collaboration, AEIL is committed to advancing both artificial intelligence and engineering to create more resilient and intelligent solutions.
AEIL’s research is firmly grounded in the principles of advancing AI integration into technical systems through interdisciplinary collaboration, with a keen emphasis on explainable AI, physics-informed neural networks, hyperparameter tuning, simulation, meta-models, and optimization. By collaborating with technologists, scientists, and industry experts, AEIL employs cutting-edge methodologies to enhance the performance and accountability of AI technologies within engineering and scientific realms.
Our research domain draws inspiration from the philosophy of science, particularly „The New Experimentalism,“ a concept championed by esteemed researchers like Ian Hacking and Deborah Mayo. This framework underscores the critical of experimentation in scientific discovery and comprehension. As articulated by Leciejewski, the integration of computer-aided experiments into the new experimentalism represents a significant, yet underexplored, dimension within the philosophy of science. Developing this field further will expand the new experimentalism into a comprehensive philosophy of experimentation for the 21st century.
I am aware that there is a number of analyzes relating to computer experiments (computer simulations), which, under certain assumptions, could be considered an extension of the concept of the new experimentalism (Bartz-Beielstein, 2005). It seems, however, that the methodological and epistemological aspects of incorporating digital elements into the experimental system are still an important and unrecognized research field of the philosophy of science. Their development would allow to expand the new experiment to such an extent that it could be a philosophy of experiment of the 21st century and not just a historical concept dating back to the end of the 20th century.
In this innovative ecosystem, AEIL develops and disseminates technical knowledge, aligning research initiatives with the evolving landscape of machine technologies and their application across various technical systems. We prioritize transparency and interpretability, ensuring that our solutions are both understandable and trustworthy to stakeholders.
Physics-informed neural networks are pivotal to our mission, incorporating domain-specific knowledge into AI models, enhancing their accuracy and applicability in complex environments. By refining these models through hyperparameter tuning, we ensure their alignment with specific technical requirements, optimizing their performance.
By utilizing simulation and meta-models, AEIL enhances the experimentation and optimization of technical systems, embodying the practical experimentation ethos of „The New Experimentalism.“ This approach diminishes the need for costly physical prototypes, accelerates innovation, and fosters the development of more advanced solutions.
The core mission of the Artificial and Engineering Intelligence Lab is to elevate interdisciplinary scholarship within the technical community, nurturing dialogue and collaboration among experts, students, and policymakers globally. Through informed decision-making and a steadfast commitment to empirical research, AEIL plays a crucial in advancing the research, implementation, and continuous evolution of cutting-edge digital technologies within the technical realm.
References
Ackermann, R. The New Experimentalism. British Journal for the Philosophy of Science 40 (1989), 185–190.
Bartz-Beielstein, T. The New Experimentalism. In: Experimental Research in Evolutionary Computation. Natural Computing Series. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-32027-X_2
Chalmers, A. F. What Is This Thing Called Science. University of Queensland Press, St. Lucia,
Australia, 1999.
Hacking, I. Representing and Intervening. Cambridge University Press, Cambridge, U.K., 1983.
Leciejewski, S. G. (2023). New experimentalism and computer-aided experiments. Philosophical Problems in Science (Zagadnienia Filozoficzne W Nauce), (75), 107–134. https://doi.org/10.59203/zfn.75.641
Mayo, D. G. Error and the Growth of Experimental Knowledge. The University of Chicago Press,
Chicago IL, 1996.