data_UMR
Aaron Beller
Welcome to data_UMR!
Data_UMR is a cross faculty publication hub. We collect scientific resources and research data from members of Philipps-University Marburg and make it openly accessible. To ensure high standards of quality and re-usability, submissions to data_UMR are subject to curation.
Select a community to browse its collections.
Recent Submissions
Distracted by Previous Experience: Integrating Selection History, Current Task Demands and Saliency in an Algorithmic Model- Springer2024
(Philipps-Universität marburg, 07.05.2021) Endres, Dominik; Meibodi, Neda; Meibodi, Neda
This repository contains the files and data necessary to recreate the results from the paper Meibodi, N., Abbasi, H., Schubö, A. et al. Distracted by Previous Experience: Integrating Selection History, Current Task Demands and Saliency in an Algorithmic Model. Comput Brain Behav 7, 268–285 (2024). https://doi.org/10.1007/s42113-024-00197-6
Please go to Version 2 if you are interested to see the files related to the other pubplication N.Meibodi, H.Abbasi, A. Schuboe, D. Endres (2021) A Model of Selection History in Visual Attention, Proceedings of the 2021 Conference of the Society of Cognitive Science, Vienna, Austria.
Recognition of European mammals and birds in camera trap images using deep neural networks
(Philipps-Universität Marburg) Schneider, Daniel; Lindner, Kim; Vogelbacher, Markus; Bellafkir, Hicham; Mühling, Markus; Farwig, Nina; Freisleben, Bernd
This record contains the trained models and the test data sets presented in the papers "Recognizing European mammals and birds in camera trap images using convolutional neural networks" (Schneider et al, 2023) and "Recognition of European mammals and birds in camera trap images using deep neural networks" (Schneider et al., 2024). In these publications, we present deep neural network models to recognize both mammal and bird species in camera trap images.
In the archive files "model2023_ConvNextBase.tar" and "model2023_EfficientNetV2.tar" as well as "model2024_ConvNextBase_species.tar" and "model2024_ConvNextBase_taxonomy.tar" we provide downloads of the best trained models from our 2023 and 2024 papers, respectively. All models are provided in the Tensorflow2 SavedModel format (https://www.tensorflow.org/guide/saved_model). A script to load and run the models can be found in our Git-Repository: https://github.com/umr-ds/Marburg-Camera-Traps. There we also provide a code snippet to perform predictions with these models.
In the archive files "data_MOF.tar" and "data_BNP.tar", we provide downloads for our Marburg Open Forest (MOF) and Białowieża National Park (BNP) data sets, consisting of about 2,500 and 15,000 labeled camera trap images, respectively. The files contain two folders named "img" and "md", respectively. The "img" folder contains the images grouped in subfolders by recording date and camera trap id. The "md" folder contains the metadata for each image, which constists of the bounding box detections obtained using the MegaDetector model (https://github.com/agentmorris/MegaDetector). The metadata is grouped into yaml-files for each label at different taxonomic levels.
Uncertainty of treatment efficacy moderates placebo effects on reinforcement learning
(Philipps-Universität Marburg) Augustat, Nick; Endres, Dominik; Müller, Erik Malte
This repository includes code and data necessary for running the online survey and task, and the analyses of the manuscript "Augustat, N., Endres, D., Müller, E. M.: Uncertainty of treatment efficacy moderates placebo effects on reinforcement learning".
Abstract: The placebo-reward hypothesis postulates that positive effects of treatment expectations on health (i.e., placebo effects) and reward processing share common neural underpinnings. Moreover, experiments in humans and animals indicate that reward uncertainty increases striatal dopamine, which is presumably involved in placebo responses and reward learning. Therefore, treatment uncertainty, analogously to reward uncertainty, may affect reward learning after placebo treatment. Here, we address whether different degrees of uncertainty regarding the efficacy of a sham treatment affect reward learning. In an online between-subjects experiment with N=141 participants, we systematically varied the provided efficacy instructions before participants first received a sham treatment that consisted of listening to binaural beats and then performed a probabilistic reinforcement learning task. We fitted a Q-learning model including two different learning rates for positive (gain) and negative (loss) reward prediction errors and an inverse gain parameter to behavioral decision data in the reinforcement learning task. Our results yielded an inverted-U-relationship between provided treatment efficacy probability and learning rates for gain, such that higher levels of treatment uncertainty, rather than of expected net efficacy, affect presumably dopamine-related reward learning. These findings support the placebo-reward hypothesis and suggest harnessing uncertainty in placebo treatment for recovering reward learning capabilities.
Processing unexpected social feedback in depression
Kirchner, Lukas