Summerschool 2025: Deep learning for biodiversity and ecological research

Start: Aug 25, 2025 12:00 AM (Local Time Germany)
End: Aug 29, 2025 12:00 AM
Location: iDiv Leipzig

The participants will be introduced to the theoretical foundations of deep learning and its applications in ecological research. In addition, hands-on sessions will allow participants to deepen their understanding by working with real-world examples.The summerschool will be performed by Professor Patrick Mäder (TU Ilmenau), Dr Jana Wäldchen (MPI for Biogeochemistry) and Dr. Ladislav Hodac (MPI for Biogeochemistry).The summer school is aimed at MSc students and doctoral researchers.

Didactic aims and elements

Participants will acquire knowledge in diverse fields.

Technical deep learning skills:

Ability to use deep learning frameworks (e.g., TensorFlow, Keras, PyTorch) to build and train models.
Understanding of core deep learning concepts like neural networks, backpropagation, and activation functions.
Experience in training models on ecological data (e.g., species classification, trait recognition).

Data handling and preprocessing:

Competence in preprocessing ecological data (e.g., labeling, cleaning, normalization, augmentation) for deep learning tasks.
Experience in working with structured and unstructured data, including images, sensor data, and field observations.

Model evaluation and optimization:

Skills in evaluating model performance using appropriate metrics (accuracy, precision, recall, etc.) and optimizing deep learning models to improve results.

Ecological applications of AI:

Knowledge of how deep learning can be applied to key biodiversity issues, such as monitoring invasive species, tracking phenology and identifying species from images.