📢Several PhD and Postdoc positions in my group (focusing on AI for Scientific Discovery in Physics) at @unituebingen.bsky.social

If you are excited about using AI to design new ways for observing the universe, please apply here: uni-tuebingen.de/en/128980#c2...

Had great fun in my first ever, live, radio interview, with ORF's Ö1 (Punkt Eins, with Barbara Zeithammer) -- on the occasion of my new professorship at the University of Tübingen @ml4science.bsky.social, talking about AI for Science.

Listen to it here: oe1.orf.at/programm/202...
(54mins, german)

The humans behind the Artificial Scientist Lab @mpi-scienceoflight.bsky.social --> @unituebingen.bsky.social 🤘

We show how to estimate the impact of research ideas before they were even born -- just published in #MLST

[paper] iopscience.iop.org/article/10.1...
[GitHub]: github.com/artificial-s...
spearheaded by @xmgu.bsky.social

Important for future artificial muses, to generate impactful new ideas.

Very happy about our super-fast JAX simulator for optics, with an application in automated discovery of new microscopes. 🔬🤖🧠

Spearheaded by Carla Rodriguez, /w Soeren Arlt & domain expert Leonhard Moeckl. Fantastic collab made possible by the Max Planck environment!

www.nature.com/articles/s41...

Very happy about this new PRL paper! 🥳

It's an experiment of an AI-discovered idea that was overlooked for 25+ years:

An alternative to entanglement swapping for entangling two independent photons - without (!) using entanglement or Bell state measurements.

Paper: journals.aps.org/prl/abstract...

Depends on the loss of the system, without loss, about 180Hz for 4-photon emitter simulation, with realistic loss, about 40% of that. This seems to be higher than experiments and proposals for 3-photon emitters that we found.

If a clever algorithm finds a super-human solution: What did the machine discover that humans have overlooked?

Happy to have one answer for a super-human quantum optics solution published in IOP's #Quantum Science & Technology: iopscience.iop.org/article/10.1...
by Soeren Arlt & Carlos Gonzalez

Happy to see our work on boosting creative idea-generation for scientists mentioned in DeepMind's text.

See the corresponding paper, on SciMuse spearheaded by @xmgu.bsky.social:

arxiv.org/abs/2405.17044

One thing is clear: the future will be exciting :)

Had the opportunity to visit the Wendelstein 7X nuclear fusion reactor in Greifswald, Germany.

It is a Stellerator & the magnetic confinement design were only possible with computers in the 90s.

The control room looks like one for NASA's rocket launches - absolutely amazing science & technology!

My favorite conclusion: Using the 4,000 human-expert evaluations, we can now predict and improve interest-levels of ideas! Thus can generate much more targeted suggestions in the future.

The future will be wild :)

Towards an Artificial Muse for Science, we want to know how interesting AI-generated research ideas are: arxiv.org/abs/2405.17044

100+ Research Group Leaders (!) from 50+ Max Planck Institutes helped us evaluate! Spearheaded by
@xmgu.bsky.social

Finally received my copy of this book again (after reading it 15y ago).

Artificial Life studies evolutionary processes of self-replicating programs in virtual chemistries, such as Tierra or avida (pioneered by Tom Ray & Christoph Adami).

Allows for open-ended evolution without an explicit goal!