- Veranstaltungsort
- Online
- Art
- Tagung
Auch in diesem Jahr gibt es ein spannendes und abwechslungsreiches Programm rund um das Thema der Künstlichen Intelligenz.
Register here to the symposium - coming soon | Hier können sie sich demnächst für das Symposium registrieren
Das englischsprachige Symposium ist kostenlos und auch für externe Teilnehmer (ohne KI-Vorwissen) zugänglich. Geplant ist ein ganzer Tag mit verschiedenen Themen, die im weitesten Sinne mit KI und Lebenswissenschaften zu tun haben.
Key Note: Prof. Dr. Karsten Borgwardt, Max-Planck-Institut |
Sebastian Burkhart, HSWT In this study, we explore the impact of the Train-Split-Ratio (TSR) on spectral grain yield prediction using AI, analyzing overall seven field trials from two locations over three and four years. We compare six algorithms across a TSR range from 5% to 95%, evaluating their performance based on a red edge vegetation index. Additionally, we investigate the optimization of neural network architectures to enhance prediction accuracy. Our findings provide insights into the optimal balance of training and test data and effective neural network designs for yield forecasting in agricultural trials. |
Andreas Gilson, Fraunhofer IIS This talk will present FruitNeRF as part of the For5G project that has the overall goal of creating of end-to-end pipelines for digital twins in horticulture. FruitNeRF is a novel unified fruit counting framework that leverages state-of-the-art view synthesis methods to count any fruit type directly in 3D. Utilizing neural radiance fields and the foundation model SAM, it becomes possible to count arbitrary types of fruit based on an unordered set of 2D images without extensive manual labeling. The presented method also prevents typical pitfalls in fruit counting, like double counting or counting of fallen or background fruit and was evaluated on synthetic and real-world datasets. |
Lars Kappertz, Center for Industrial Mathematics Modelling and Optimal Control of Thermal Storages in a Smart Energy Management System Forecast-based energy management can play a large role in a smarter and more efficient use of renewable energies based on demand side management. Using approaches such as model predictive control, individual consumption devices can be shifted within operation constraints so that their electricity consumption optimally matches generation. In agriculture, large thermal storages make up a sizeable part of electricity consumption, and offer a potential use in the short term shifting of demand. Necessary for this are accurate models to forecast behaviour of such dynamic systems, so that minimal power demand and fulfilment of operation constraints can be ensured when computing optimal controls. |
Christine Drießlein, HSWT Analysis of the combination suitability between different NMR metabolite profiles using artificial intelligence methods This project aims to establish a connection between certain properties of dandelion species, such as high rubber content, and their metabolite profiles using multivariate and machine learning methods. The calculated models shall help to gain insights into relevant individual metabolites and metabolite networks and, thus, to understand the underlying biochemical mechanism [1]. The metabolite profiles required for the analysis are calculated automatically from 1H NMR spectra of the dandelion plants with the help of a self-written computer program. The significance of this work lies in the potential for dandelions to become an alternative source of natural rubber production [2], mitigating environmental concerns associated with traditional rubber production, while creating significant regional value chains. So far, only molecular approaches have been pursued to identify relevant genetic markers in dandelions for rubber content and root morphology [2], while the metabolome has not been considered. Plant material, from leaves and roots, underwent optimized sample preparation and one-dimensional 1H NMR measurements were conducted using a 600 MHz Bruker NMR spectrometer. Metabolites were automatically identified and quantified from spectra using a self-written identification algorithm, non-linear optimization methods and an extensive database. No precise details can be given about the further data analysis, as these studies had just begun at the time of submission. As part of this project, 142 metabolites were measured in dandelion matrices. Among these, 34 known metabolites were confirmed, while 22 new metabolites were identified. Together, these 56 metabolites account for most signals in dandelion spectra, while the remaining identified metabolites are present in smaller concentrations and thus intensity. Automated identification processes demonstrated high accuracy, providing a basis for subsequent analyses. Initial results of statistical evaluations are presented and their significance for a deeper understanding of the underlying biochemical processes is discussed. This project represents a crucial step towards understanding the biochemical mechanisms underlying rubber yield enhancement in dandelion. The use of advanced data analysis methods highlights the importance of continued research in this field for further advancements in biotechnology and sustainable agriculture. References [1] C. Riedelsheimer, et al, Nat Genet 44, 217–220 (2012). [2] A. Stolze, et al., Plant Biotechnol J. 15, 740-753 (2017). |
Judith Bernett, TUM School of Life Sciences How data leakage hinders real progress in the field of PPI prediction
References
|
Die Einladung zum KI-Symposium 2024 finden Sie demnächst hier
Das waren die letztjährigen Veranstaltungen:
KI-Symposium am 20. Oktober 2023
Key Note-Speaker: Dr. Martin Junghans, CTO Innovation Studios, IBM Deutschland GmbH
KI-Symposium am 21. Oktober 2022
Key Note-Speaker: Prof. Dr. Joachim Hertzberg (Universität Osnabrück)
KI-Symposium am 22. Oktober 2021
Key Note-Speaker: Jonas Andrulis (Aleph Alpha GmbH) und Univ.-Prof. Dr. Sepp Hochreiter (Johannes Kepler Universität Linz), der den diesjährigen KI-Innovationspreis der WELT erhalten hat.