@sbluecher

PhD @ml_tuberlin and @bifoldberlin Physics graduate • Explainable AI • Kernel learning

Joined January 2022
  • Tweets26
  • Following147
  • Followers64
2Media
Stefan Blücher retweeted
BIFOLD@bifoldberlin6 Jan 2025
Join BIFOLD as a doctoral student in our Graduate School. Apply until February 03, 2025. Job ID: jobs.tu-berlin.de/en/job-pos… #jobvacancy #jobalert #ScienceJob #AI #research #Berlin
4
2
4
2,900
Our work on historical insights at scale using machine learning is now out in @ScienceAdvances! Very proud of this team effort, bridging disciplines and institutions—@MPIWG @TUBerlin @bifoldberlin @ml_tuberlin 📜science.org/doi/10.1126/scia…
3
15
3
29
5,090
Stefan Blücher retweeted
BIFOLD@bifoldberlin11 Oct 2024
#Job for #PhDCandidates: Develope and implement #XAI methods for ML models in Quantum Chemistry. BIFOLD is seeking a Research Associate in #ML for a project supervised by Dr. Stefan Chmiela and Dr. Shinichi Nakajima. Apply by 7th Nov '24 linkedin.com/jobs/view/40470… @ml_tuberlin
Independent and responsible research in the field of machine learning. The goal of the advertised project is the development of new explanation methods (Explainable Artificial Intelligence, XAI) for atomistic modeling in quantum chemistry. The focus is on acquiring physically grounded insights to guide hypotheses in quantum chemistry.

ALT Independent and responsible research in the field of machine learning. The goal of the advertised project is the development of new explanation methods (Explainable Artificial Intelligence, XAI) for atomistic modeling in quantum chemistry. The focus is on acquiring physically grounded insights to guide hypotheses in quantum chemistry.

1
1
4
2,228
Stefan Blücher retweeted
Nils Strodthoff@nstrodt7 Oct 2024
I have two vacant PhD positions in my group: 🏃 Machine Learning for Activity Trajectories s.gwdg.de/cSWx1D (Deadline Oct 20) 🧪 Machine learning for Medical Diagnostics s.gwdg.de/6NJfae (Deadline Oct 31) Looking forward to your applications!
1
6
5
1,061
Stefan Blücher retweeted
Julius Hense@hense9630 Sep 2024
✨ Our xMIL paper has been accepted at #NeurIPS2024 ✨ arXiv preprint: arxiv.org/abs/2406.04280
2
4
2
15
2,202
Stefan Blücher retweeted
Konrad Rieck 🌈@mlsec16 Sep 2024
🚀Last-minute check-in to an academic career? We're seeking a PhD student to research the security and privacy of generative AI. Fully funded position in vibrant Berlin! Deadline is this Friday, September 20th. Apply now: mlsec.org/jobs.html#jobs @bifoldberlin @TUBerlin

Jobs at Chair of Machine Learning and Security

We are looking for motivated PhD students and postdocs to join our research group at TU Berlin.

mlsec.org
12
1
24
2,859
🔊 New publication in @TmlrOrg Decoupling Pixel Flipping and Occlusion Strategy for Consistent #XAI Benchmarks, by @sbluecher @jvielhaben and @nstrodt together with @bifoldberlin and @FraunhoferHHI 🙏🏼 #MachineLearning #AI
Stefan Blücher@sbluecher8 Jul 2024
Greatful to be published @TmlrOrg. Thanks to @nstrodt and @jvielhaben for their support. @bifoldberlin @ml_tuberlin @FraunhoferHHI
1
4
11
1,307
Greatful to be published @TmlrOrg. Thanks to @nstrodt and @jvielhaben for their support. @bifoldberlin @ml_tuberlin @FraunhoferHHI
Accepted papers at TMLR@TmlrPub6 Jul 2024
Decoupling Pixel Flipping and Occlusion Strategy for Consistent XAI Benchmarks Stefan Bluecher, Johanna Vielhaben, Nils Strodthoff. Action editor: Wei Liu. openreview.net/forum?id=bIiL… #occlusion #features #feature
1
1
9
1,632
Decoupling Pixel Flipping and Occlusion Strategy for Consistent XAI Benchmarks Stefan Bluecher, Johanna Vielhaben, Nils Strodthoff. Action editor: Wei Liu. openreview.net/forum?id=bIiL… #occlusion #features #feature
3
2
8
2,471
Stefan Blücher retweeted
Nils Strodthoff@nstrodt7 Jul 2024
Our work on pixel flipping methodologies for #XAI methods got accepted at TMLR. Give our SRG measure a try the next time you benchmark your latest XAI method. Thanks @sbluecher @jvielhaben for the fun collaboration @XAI_Research @FraunhoferHHI @bifoldberlin
Accepted papers at TMLR@TmlrPub6 Jul 2024
Decoupling Pixel Flipping and Occlusion Strategy for Consistent XAI Benchmarks Stefan Bluecher, Johanna Vielhaben, Nils Strodthoff. Action editor: Wei Liu. openreview.net/forum?id=bIiL… #occlusion #features #feature
4
13
1,187
Stefan Blücher retweeted
BIFOLD@bifoldberlin20 Jun 2024
👩‍🔬Meet some of the BIFOLD researchers of the KI-Schnitzeljagd! Learn about AI while having fun. Don’t miss it! June 22, 2024 Fraunhofer HHI, Berlin langenachtderwissenschaften.… #BehindTheScenes #AIResearch #LNDW2
1
2
309
Stefan Blücher retweeted
BIFOLD@bifoldberlin3 Jun 2024
Join BIFOLD and Machine Learning Group at @TUBerlin Berlin as a #PhD candidate. Conduct cutting-edge research in deep learning & explainable AI. Apply by June 14, 2024. jobs.tu-berlin.de/en/job-pos… @ml_tuberlin
2
5
8
3,146
Interested in higher-order #XAI Check out my recent package: github.com/bluecher31/hoocs It allows to calculater arbitrary interactions of features for #PredDiff and Shapley values. @trustworthy_ml @XAI_Research

GitHub - bluecher31/hoocs

Contribute to bluecher31/hoocs development by creating an account on GitHub.

github.com
2
1
8
720
Stefan Blücher retweeted
BIFOLD@bifoldberlin23 Jan 2024
Stefan Blücher is a member of the BIFOLD Graduate School. The next application deadline is closing soon: 26. Jan 2024. bifold.berlin/about-us/oppor… 10 open positions #PhD #ResearchJob #Berlin #ScienceJob xitter.cf/bifoldberlin/status/17… – at Berlin, Germany
BIFOLD@bifoldberlin20 Dec 2023
Replying to @bifoldberlin
🚀10 open positions, Research Assistant, E13 👨‍💻BIFOLD Graduate School 🔥Deadline: Jan 26, 2024 🔗 t1p.de/ix48u #PhDcandidates Graduate School Infos: bifold.berlin/education/grad… #sciencejobs #research #job
© BIFOLD / Welcome event October 2022: BIFOLD Graduate School members and team. The Graduate School educates students for the rising worldwide demand for specialists with expertise in Data Management and Machine Learning. It is highly research-focused and strives to build the next generation of curious and creative data science experts who challenge assumptions, find answers to significant questions, and exercise ethical responsibility. Based on the overarching research foci of BIFOLD, the Graduate School is offering PhD projects in the areas of current challenges in Artificial Intelligence and Data Science, focusing on data management, machine learning, and their intersection. To learn more about recent research, please visit the research groups' detailed pages

ALT © BIFOLD / Welcome event October 2022: BIFOLD Graduate School members and team. The Graduate School educates students for the rising worldwide demand for specialists with expertise in Data Management and Machine Learning. It is highly research-focused and strives to build the next generation of curious and creative data science experts who challenge assumptions, find answers to significant questions, and exercise ethical responsibility. Based on the overarching research foci of BIFOLD, the Graduate School is offering PhD projects in the areas of current challenges in Artificial Intelligence and Data Science, focusing on data management, machine learning, and their intersection. To learn more about recent research, please visit the research groups' detailed pages

1
266
Faithful attributions are measured by pixel flipping! But how trustworthy and consistent are the PF benchmarks? Read our new preprint to learn more! @nstrodt @jvielhaben @ml_tuberlin @bifoldberlin @FraunhoferHHI #XAI #diffusion arxiv.org/abs/2401.06654

Decoupling Pixel Flipping and Occlusion Strategy for Consistent...

Feature removal is a central building block for eXplainable AI (XAI), both for occlusion-based explanations (Shapley values) as well as their evaluation (pixel flipping, PF). However, occlusion...

arxiv.org
1
4
2
11
1,611
(1) grouping PF setups according to the R-OMS score clusters consistent MIF and LIF rankings. (2) decoupling reliability and insightfulness of PF benchmarks by combing MIF and LIF into the 'Symmetric Relevance Gap' measure.
110
🎙️ In todays #PhD seminar @sbluecher and Marco Morik presented their #research on „Assessing what really matters for occlusion-based #XAI“ and „#DiffusionModels for Inverse Problems“ Thanks guys for the interesting talks and discussions! #MachineLearning
1
2
10
566
Follow our lab to stay tuned for our most recent research highlights! #MachineLearning @bifoldberlin @TUBerlin
ML Group, TU Berlin@ml_tuberlin14 Jun 2023
This is the #MachineLearning Group of Prof. Klaus-Robert Müller at the @TUBerlin. Let’s connect! #ScienceTwitter
1
8
322
🎉🥳 Greatful to share that our recent research on concept-based explanations got expected @TmlrPub. @nstrodt (@UniOldenburg) and Johanna Vielhaben (@FraunhoferHHI). @bifoldberlin #XAI
Accepted papers at TMLR@TmlrPub5 Jun 2023
Multi-dimensional concept discovery (MCD): A unifying framework with completeness guarantees Johanna Vielhaben, Stefan Bluecher, Nils Strodthoff. Action editor: Hsuan-Tien Lin. openreview.net/forum?id=KxBQ… #concepts #concept #clustering
2
2
10
1,598
Stefan Blücher retweeted
Nils Strodthoff@nstrodt6 Jun 2023
🎉🎊🎉 I am happy to announce that our work on multi-dimensional concept discovery #XAI method was accepted @TmlrPub openreview.net/forum?id=KxBQ… Be sure to also check out Johanna's beautiful explainer video youtube.com/watch?v=JZOzo-K0… @trustworthy_ml @XAI_Research
Nils Strodthoff@nstrodt31 Jan 2023
📢 I am excited to share our latest preprint on concept-based #XAI arxiv.org/abs/2301.11911 (joint work with @sbluecher from @TUBerlin and J Vielhaben @FraunhoferHHI ) @trustworthy_ml @XAI_Research

Multi-dimensional concept discovery (MCD): A unifying framework...

The completeness axiom renders the explanation of a post-hoc XAI method only locally faithful to the model, i.e. for a single decision. For the trustworthy application of XAI, in particular for...

arxiv.org
1
5
247
Load more