Stefan Michaelis

Stefan Michaelis

@stmichaelis.bsky.social

@Lamarr Institute for Machine Learning and Artificial Intelligence Sharing insights, news, and findings from the web https://lamarr-institute.org/

49 Followers 77 Following 108 Posts Joined Nov 2024
1 day ago
Preview
Lamarr Exhibit at MS Wissenschaft » Lamarr News A Lamarr Institute exhibit on the MS Wissenschaft demonstrates how AI and digital twin models analyze causal relationships in medical data, enabling more personalized treatment decisions and advancing...

News from Lamarr: "Lamarr Exhibit at MS Wissenschaft" at:

0 1 0 0
1 week ago
Anti-patterns: things to avoid - Agentic Engineering Patterns Anti-patterns: things to avoid - Agentic Engineering Patterns

Recommended reading: "Anti-patterns: things to avoid" at:

0 0 0 0
2 weeks ago
Preview
AI slop It’s not slop because it was created by an AI. It’s slop because it’s slop. I just read the first two pages of a sci-fi novel on my Kindle. The author proudly proclaims that the 4…

Recommended reading: "AI slop" at:

0 0 0 0
3 weeks ago
Preview
A Guide to Which AI to Use in the Agentic Era It's not just chatbots anymore

Recommended reading: "A Guide to Which AI to Use in the Agentic Era" at:

0 0 0 0
3 weeks ago
Preview
Process thinking Sure, you made it work this time, but will it work next time? Can you teach the method to someone else? Do you have a protocol for what to do when it doesn’t work? How can someone else contri…

Recommended reading: "Process thinking" at:

0 0 0 0
3 weeks ago
Preview
25th Schnupper Uni at the University of Bonn » Lamarr News At the University of Bonn’s Schnupper Uni, Prof. Lena Funcke inspired 225 female students with insights into AI research, theoretical physics, and pathways into STEM careers.

News from Lamarr: "25th Schnupper Uni at the University of Bonn" at:

0 0 0 0
1 month ago
Preview
AI Guardrails Do Not Work (Yet) Would you cross a bridge that was 99.7% safe? The answer is not necessarily simple. It depends on how many times you need to cross it (once, or every day for...

Recommended reading: "AI Guardrails Do Not Work (Yet)" at:

0 0 0 0
1 month ago
Preview
AALearning: Uncertainty-Aware AI for Fundamental Physics AALearning develops uncertainty-aware AI methods to improve robustness and reliability in fundamental physics research.

News from Lamarr: "AALearning project launches to advance safe and reliable AI in fundamental physics" at:

0 0 0 0
1 month ago
Preview
AI Must Be Line Managed I used to just code. Then I managed teams. Now I find myself needing management skills again, even when I am not managing people.The first time I tried deleg...

Recommended reading: "AI Must Be Line Managed" at:

0 0 0 0
1 month ago
Preview
New cooperation strengthens international AI research at Lamarr » Lamarr News The Lamarr Institute, the University of Bonn, and the Nara Institute of Science and Technology (NAIST) have launched a long-term research collaboration in artificial intelligence, life sciences, and data...

News from Lamarr: "New cooperation strengthens international AI research at Lamarr" at:

2 0 0 0
1 month ago
Preview
Scaling laws of emergent multilingualism in vision-language models » Lamarr News New research identifies scaling laws that explain how vision-language models develop emergent multilingual image captioning abilities through generalization from translation data, even without explicit...

News from Lamarr: "Scaling laws of emergent multilingualism in vision-language models" at:

0 0 0 0
1 month ago
Preview
Prof. Dr. Julia Tjus elected to the Royal Swedish Academy of Sciences » Lamarr News Prof. Dr. Julia Tjus has been appointed as an international member of the Royal Swedish Academy of Sciences. Her work at the interface of theoretical physics and artificial intelligence strengthens Lamarr’s...

News from Lamarr: "Prof. Dr. Julia Tjus elected to the Royal Swedish Academy of Sciences" at:

0 0 0 0
1 month ago
Preview
Physics-LLM: Advancing from Big Data to Smart Data » Lamarr News Funded by the German Federal Ministry of Research, Technology and Space (BMFTR), Physics-LLM is an ErUM-Data research project developing AI-based tools using large language models to improve research...

News from Lamarr: "Physics-LLM: Advancing from Big Data to Smart Data" at:

0 0 0 0
2 months ago

Recommended reading: "A field guide to sandboxes for AI" at:

0 0 0 0
2 months ago
Preview
AprielGuard: A Guardrail for Safety and Adversarial Robustness in Modern LLM Systems A Blog post by ServiceNow-AI on Hugging Face

Recommended reading: "AprielGuard: A Guardrail for Safety and Adversarial Robustness in Modern LLM Systems" at:

0 0 0 0
2 months ago
Preview
Lamarr researcher Prof. Dr. Petra Wiederkehr elected to the German Academy of Science and Engineering (acatech) » Lamarr News Lamarr Principal Investigator Petra Wiederkehr has been elected to Germany’s National Academy of Science and Engineering (acatech).

News from Lamarr: "Lamarr researcher Prof. Dr. Petra Wiederkehr elected to the German Academy of Science and Engineering (acatech)" at:

0 0 0 0
2 months ago
Preview
Benchmark Studies It is impossible to disentangle technical innovation from technical debt

Recommended reading: "Benchmark Studies" at:

0 0 0 0
2 months ago
Preview
True Stories from the (Data) Battlefield – Part 1: Communicating About Data A blog about data science, statistics, and data analysis with open-source software.

Recommended reading: "True Stories from the (Data) Battlefield – Part 1: Communicating About Data" at:

0 0 0 0
3 months ago
Preview
ChatGPT Is Terraforming Earth Your ChatGPT subscription is subsidising the terraforming of Earth.Not metaphorically. China’s solar overbuild is crashing electricity prices toward zero. AI...

Recommended reading: "ChatGPT Is Terraforming Earth" at:

0 0 0 0
3 months ago
A quote from Obie Fernandez If the part of programming you enjoy most is the physical act of writing code, then agents will feel beside the point. You’re already where you want to be, even …

Recommended reading: "Quoting Obie Fernandez" at:

0 0 0 0
3 months ago
The Normalization of Deviance in AI This thought-provoking essay from Johann Rehberger directly addresses something that I’ve been worrying about for quite a while: in the absence of any headline-grabbing examples of prompt injection...

Recommended reading: "The Normalization of Deviance in AI" at:

0 0 0 0
3 months ago
Preview
The Slow And Dangerous Loss Of Self I have been telling people to use AI as a thinking partner, not a replacement for thinking. To let it interview them, surface ideas, enhance their output rat...

Recommended reading: "The Slow And Dangerous Loss Of Self" at:

0 0 0 0
3 months ago
Preview
RePAIR brings Pompeii's frescoes back to life » Lamarr News The European research project RePAIR – Reconstructing the Past combines robotics, AI, and computer vision to reconstruct ancient frescoes from Pompeii using an AI-supported method that digitally analyzes...

News from Lamarr: "RePAIR brings Pompeii’s frescoes back to life" at:

0 0 0 0
3 months ago
Preview
Award for sustainable AI research » Lamarr News Award-winning research on sustainable large language models recognized by TU Dortmund for advancing energy-efficient and transparent AI systems.

News from Lamarr: "Award for sustainable AI research" at:

1 0 0 0
3 months ago
Preview
Data spaces and vertical AI: Europe's path to digital sovereignty » Lamarr News Vertical AI and trusted data spaces emerge as essential pillars for Europe’s digital sovereignty, highlighted by Lamarr Institute in Paris

News from Lamarr: "Data spaces and vertical AI: Europe’s path to digital sovereignty" at:

2 0 0 0
3 months ago
Preview
New access to quantum computing » Lamarr News Open-access textbook presenting an intuitive quantum computing introduction by bridging AI fundamentals, Hopfield networks, and practical Python examples

News from Lamarr: "New access to quantum computing" at:

2 0 0 0
3 months ago
Preview
Setting the course for European AI sovereignty » Lamarr News European AI sovereignty takes center stage as the Foundation Model Workshop contributes research perspectives to the Summit on European Digital Sovereignty

News from Lamarr: "Setting the course for European AI sovereignty" at:

0 0 0 0
3 months ago
Preview
JUPITER selects first AI projects » Lamarr News Bonn’s Lamarr Institute wins JUPITER Gauss-AI selection: Prof. Gall’s HoMe project advances generative AI, robotics, and human-motion video models.

News from Lamarr: "JUPITER selects first AI projects" at:

0 0 0 0
4 months ago
Preview
Lamarr Institute at the Cine Bonn Bleu Short Film Festival Experts from design, film, and technology discussed the creative potential & ethical challenges of AI at Cine Bonn Bleu Short Film Festival moderated by Lamarr.

News from Lamarr: "“AI – Muse or Monster?”: Lamarr Institute at the Cine Bonn Bleu Short Film Festival" at:

0 0 0 0
4 months ago
Preview
RiverMamba: New AI architecture improves flood forecasting » Lamarr News Artificial intelligence for better disaster prevention: The new AI model “RiverMamba”, developed at the Lamarr Institute and the University of Bonn under the lead of Prof. Dr. Jürgen Gall, significantly...

News from Lamarr: "RiverMamba: New AI architecture improves flood forecasting" at:

0 0 0 0