Welcome to the Machine Learning Lab at the University of Freiburg!
Machine learning (ML) is the key technology of our time. However, so far, its success crucially relies on human machine learning experts to perform manual tasks, thus limiting its potential impact. To overcome this issue, here at the ML Lab, we focus on the progressive automation of machine learning (AutoML) in order to democratize access to ML by making state-of-the-art ML solutions accessible for everyone, in world-leading open source systems. Framed differently, and from a technical point of view, our lab develops AI that builds and improves AI.
Particular technical fields of interest for us include meta-learning, neural architecture search, efficient hyperparameter optimization, deep learning for tabular data, multi-objective optimization, proper benchmarking, and AutoRL. We also apply AutoML to improve practical deep learning for several applications of high societal importance, such as RNA folding and design, EEG decoding, and data from the medical sector in general.
We co-organize the MOOC on AutoML at KI Campus and the lab's prof, Frank Hutter, is general chair of the AutoML conference and also the head of the ELLIS unit Freiburg, one of the 17 founding units of the European Laboratory for Learning and Intelligent Systems (ELLIS).
On this university website, we provide information about our publications, teaching, open student projects, and how to get in touch to start working with us. We discuss AutoML in detail on our project webpage AutoML.org, including our open-source libraries, our blog, upcoming and past events, and our MOOC.
| News | |
|---|---|
| 08.2024 | We will present 20 (!) papers at AutoML conference 2024 |
| 07.2024 | Our team will present 6 papers at ICML 2024: 3 at the main conference and 3 at workshops! |
| 04.2024 | We'll present 7 papers at ICLR 2024: 2 at the main conference and 5 at workshops |
| 09.2023 | We’re delighted to announce that 6 AutoML papers from our team have been accepted to the NeurIPS2023 main track, including an oral for a paper on AutoML for Trustworthy AI! We also have 2 additional workshop papers and Frank has a keynote at the Table Representation Workshop |
| 09.2023 | We are organizing the 2023 AutoML Conference in Potsdam |
| 08.2023 | Auto-WEKA received the KDD 2023 Test of Time Award for Research |
| 08.2023 | We take part in the Small Data project that aims to overcome the challenges of learning on small data |
| 07.2023 | Frank received the Award for Excellent Supervision of Doctoral Students at the University of Freiburg |
| 07.2023 | Frank nominated by the Technical Faculty for the University Teaching Award |
| 07.2023 | Our AutoML course received the Best Teaching Evaluation at the Technical Faculty |
| 01.2023 | Konrad Zuse School of Excellence in Learning and Intelligent Systems (ELIZA) research grant awarded as part of consortium of 7 German ELLIS Units, sponsored by BMBF and DAAD |
| 12.2022 | We are organizing the next AutoML Conference! All important dates and details can be found through automl.cc |
| 12.2022 | TabPFN won the best paper award at the NeurIPS'22 TRL Workshop |
| 11.2022 | We'll present 17 papers at NeurIPS 2022: 2 at the main conference, 2 at the datasets and benchmark track, and 13 at workshops! |
| 10.2022 | We're organizing the AutoML Fall School 2022 here in Freiburg |
| 07.2022 | Our GECCO’22 paper Theory-Inspired Parameter Control Benchmarks for Dynamic Algorithm Configuration won the best paper award on the GECH track |
| 07.2022 | Frank is the co-initiator and inaugural general chair of the first AutoML conference |
| 07.2022 | We won the first prize of the AutoML 2022 Competition on Multiobjective Optimization for Transformers |
| 05.2022 | With a consortium of 10 partners, we have been awarded a grant by the ZEISS Foundation for the ReScale project |
| 03.2022 | Frank has been awarded an ERC Consolidator Grant for the topic Deep Learning 2.0! |
| 2021 | Frank was appointed as ELLIS Director for the Freiburg Unit |
| 2021 | Frank was elected fellow of the European Association for Artificial Intelligence (EurAI) |
| 2021 | Our work on Algorithm Runtime Prediction received the 2021 AIJ Prominent Paper Award |
| 2021 | Frank receives Best Teaching Evaluation at the Technical Faculty |
| 2021 | Frank gave a keynote talk at the Toronto Machine Learning Summit and the Applied AI Summit |
| 2020 | Frank has been awarded an ERC Proof of Concept Grant on AutoML |
| 2020 | We won the first prize in the NeurIPS 2020 Blackbox Optimization Challenge |
| 2020 | We launch the first MOOC on Automated Machine Learning |
| 2020 | Frank gave a keynote at ICAPS |
| 2020 | We joined the TAILOR network, focusing on the AutoAI work package. |
| 2020 | Renormalized Flows research grant awarded by the German Federal Ministry of Education and Research (BMBF) |
| 2019 | Frank was elected to the fellows of the European Laboratory for Learning and Intelligent Systems (ELLIS) |
| 2018 | Frank received the Google Faculty Research Award |
| 2018 | We won the 2nd ChaLearn AutoML Challenge 2017-2018 |
| 2018 | Frank gave a keynote at ISC High Performance and a tutorial on AutoML at NeurIPS |
| 2018 | Our industrial grant proposal in collaboration with Bosch & Uni Freiburg has been approved |
| 2017 | Frank received the ICLR Best Reviewer Award |
| 2017 | Frank gave a keynote on AutoML at ECML-PKDD and at the NeurIPS Symposium on Meta-Learning |
| 2016 | Frank has been awarded an ERC Starting Grant for the topic “Beyond Blackbox – Data-Driven Methods for Modelling and Optimizing the Empirical Performance of Deep Neural Networks” |
| 2015 | Auto-sklearn won the 1st ChaLearn AutoML challenge 2015-2016 |
| 2015 | SATzilla and AutoFolio won the first 3 prizes in the ICON Challenge on Algorithm Selection |
| 2015 | We worked on the project “RobDREAM” (funded by Horizon 2020 initiative) with a consortium of 7 participating institutions |
| 2014 | Frank received five awards at the IPC 2014 Planning and Learning Track |
| 2013 | Frank has been awarded the Emmy Noether Fellowship |
| 2013 | Frank’s work on High Dimensional Bayesian Optimisation received the IJCAI distinguished paper award |