The TNT is excited to announce that we are part of the Horizon Europe funded project Satellites for Wilderness Inspection and Forest Threat Tracking (SWIFTT). The project's goal is to develop powerful machine learning models as a backend for a platform which helps forest managers to detect threats to their forests in an affordable and simple manner. SWIFTT will be based on the remote sensing data provided by the Copernicus programme. The TNT's primary role will be developing a model for wildfire risk prediction at a high spatial resolution. SWIFTT's first press release can be found here.

Our paper Deep Reinforcement Learning for Autonomous Driving using High-Level Heterogeneous Graph Representations has been accepted to ICRA 2023! You can find the accepted version here. Code will be published shortly.

Using personalized medicine to combat depression—we are part of the largest German study to improve depression treatment. Using biomarkers to find individual diagnosis and therapy paths—what already works in oncology should also become possible in psychiatry. Under the leadership of the Hannover Medical School (MHH), a national research network is now being launched that aims to tailor the treatment of depression more closely to the individual patient than before. The project, entitled “Personalized, Predictive, Precise and Preventive Medicine to Improve Early Detection, Diagnosis, Therapy and Prevention of Depressive Disorders” (P4D), aims to develop individualized treatment approaches. Click here for the full press release (in German).

We are excited to announce the first international conference on automated machine learning (AutoML) in 2022, co-organized by Prof. Lindauer. The conference will be co-located with ICML, runs for 3 days and we are currently planning for in-person. Website: Webseite. The submission deadline is on Feb 24, 2022.

AutoML@ICML'21: We co-organize the AutoML workshop at the international conference on machine learning (ICML) 2021. In the last years, it was one of the biggest workshops at ICML. For more information visit http://icml2021.automl.org/.

Prizewinner: Aron Sommer, Jörn Ostermann

For his/her work: Analysis of Ship Deformation using the Backprojection Autofocus of Large Ships with Arbitrary Motion for Synthetic Aperture Radar

Award Sponsor: EUSAR 2021

Prizewinner: Difan Deng and Marius Lindauer et al.

AutoML.org & IOHprofiler, featuring the switching squirrel. After fixing a minor bug, our Switching Squirrel achieved a strong 3rd place at the black-box challenge at #NeurIPS2020 (#AutoML and Hyperparameter Optimization). On the warm-start friendly leaderboard, our Squirrel was even ranked 1st and was awarded an extra prize money. Our Squirrel combines ideas from (i) (simple) meta-learning by recognizing known configuration spaces, (ii) a diverse portfolio of acquisition functions, surrogate models and data transformations for Bayesian Optimization and (iii) differential evolution for a greedy final stage. It is the outcome of a joint collaboration between the groups from AutoML.org (tnt@Leibniz University Hannover and University of Freiburg) and IOHProfiler (Leiden University and Sorbonne University). Competition homepage

Prizewinner: Maren Awiszus, Frederik Schubert and Bodo Rosenhahn

For their work: TOAD-GAN: Coherent Style Level Generation from a Single Example

Award Sponsor: AIIDE 2020

Prizewinner: Felix Kuhnke, Lars Rumberg, Jörn Ostermann

In the FG-2020 Competition: Affective Behavior Analysis in-the-wild (ABAW) researchers from TNT achieved two strong 2nd places (for action unit and valence arousal estimation) and one 1st place for categorical emotion recognition. The goal of this competition was to extract human emotion from "in-the-wild" videos. Our work was presented at the IEEE International Conference on Automatic Face and Gesture Recognition (FG) 2020, 16-20 November 2020.

Award Sponsor: https://ibug.doc.ic.ac.uk/resources/fg-2020-competition-affective-behavior-analysis/

Prizewinner: Hendrik Müller

For his poster Entwicklung und Bau der Solarmessbox Hendrik Müller received the best poster award among all "FWJler" from the Leibniz University Hannover. The work was supervised by Marco Munderloh.