PhD opportunities at the interface machine learning/quantum physics
Joint call for PhD positions in our @dfg.de research unit "Machine learning for complex quantum states" together with @abohrdt.bsky.social @gppcarleo.bsky.social amongst others. Check it out at:
for5919.github.io/joinus/2025_...
Defects & interfaces in quantum hard-disk crystals are stable just due to purely quantum effects, unlike their classical counterparts. Our work reveals robust quantum dynamics, indicative of the recently developed concept of many-body caging:
doi.org/10.1103/gnqx...
Active quantum flocks
Flocks of animals are an archetype of collective behavior in the macroscopic world. Here, we show that flocks can also form at the quantum level with unique quantum features:
doi.org/10.1103/rd46...
Novel nonequilibrium phase of matter realizable in #Rydberg atomic systems.
Stylized image of a head containing a neural network, overlaying a gallery of images from scientific exploration (plots of curves and wave fields).
Scientific discovery relies on a cycle of observations, analysis, and coming up with new hypotheses. Can a computer mimic that cycle and explore an unknown system in this way?
We introduce #SciExplorer, which for the first time employs agentic behaviour in […]
[Original post on fediscience.org]
EPFL are recruiting a tenure track assistant proff working on the theory of condensed matter and interacting quantum matter:
www.epfl.ch/about/workin...
Feel free to reach out if you have any questions about life at EPFL/in Switzerland etc.
Call for Applications / Ausschreibung
Springboard for an international scientific career! 🧬🧪🔭⚛️🧠🌱 Call for #MaxPlanckResearchGroups launched; applications are possible until October 14, 2025 www.mpg.de/max-planck-r... #ScienceCareer
Group photo of postdocs conducting research at a Max Planck Institute
The #MaxPlanckPostdocProgram offers a guaranteed contract of at least 3 years, targeted mentoring, and career workshops. The call for applications is open now! 🚀 Take advantage of this opportunity and browse the job vacancies. www.mpg.de/en/max-planc...
New KITP Program! #AIQMATTER26
AI for Quantum Matter
Dates: 7/27/2026 - 10/8/2026
buff.ly/N7hkYC3
Interfaces are key to understand the physics of matter. Remarkably, (only) in 2D #quantum systems interfaces can undergo their own phase transition. The dynamical signatures of this #roughening we study in our latest PRL:
link.aps.org/doi/10.1103/...
Readily accessible in #Rydberg systems.
Review on solving dynamics of quantum matter by means of neural quantum states #NQS out now:
arxiv.org/abs/2506.03124
#machinelearning boosts real-time simulations in #quantum many-body systems, in particular in the challenging regime of two spatial dimensions.
We have several open, fully-funded PhD and Postdoc positions at my group, now at the University of Tübingen. Please #RT and share with potential candidates!
Generally: AI for scientific discoveries in physics 🔥.
mariokrenn.wordpress.com/wp-content/u...
Many-body cages: a novel mechanism for nonergodicity and nonequilibrium phases based on the emergence of flat bands in the many-body spectrum of constrained quantum matter. Yields disorder-free spin glasses and leads to coherent many-body Rabi-type oscillations:
arxiv.org/abs/2504.13086
Fractional diffusion without disorder: Local gauge constraints in 2D induce robust, tunable #subdiffusion dynamics. Relevant for platforms like artificial spin ice and quantum simulators aiming to realize discrete link models and emergent #gauge theories:
doi.org/10.48550/arX...
The dynamics in the 2D quantum hard-disk model exhibits nonergodic behavior just due quantum interference effects. We now show that these survive even perturbations, highlighting the unconventional constrained dynamics of the quantum hard disk model:
doi.org/10.48550/arX...
Recording the measurements in monitored quantum dynamics provides insights into the monitored system itself. #unsupervisedlearning shows that information loss due to enhanced entanglement is compensated by emergent structure in classical datasets of measurements:
doi.org/10.48550/arX...
Convolutional transformer wave functions
New type of neural quantum state enables solution of challenging quantum many-body problems with superior performance in both ground-state searches and dynamics. Check preprint at the interface between #quantum and #machinelearning:
doi.org/10.48550/arX...
Probing quantum many-body dynamics using subsystem Loschmidt echos
Experimentally accessing the Loschmidt echo reveals dynamical phase transitions & Hilbert space fragmentation in quantum gases. Promising tool for exploring non-equilibrium dynamics, check out here:
doi.org/10.48550/arX...
Quantum hard disks
Quantum counterpart of hard disks on a lattice yields unique quantum features. Quantum interference stabilizes crystals and leads to emergence of quantum many-body scars. Naturally realizable in Rydberg atomic systems, check it out here:
doi.org/10.1103/Phys...
Hello Bluesky! We're very excited to start sharing our research here, and connect with the #quantum #matter community ⚛️ Say hello below if you work on ultracold physics, or let us know who to follow here!
Roughening dynamics in 2D quantum matter
Interfaces are key to understand the physics of matter. In our latest preprint we study the dynamical signatures of the interface roughening transition in the 2D quantum Ising model:
doi.org/10.48550/arX...
Readily accessible in Rydberg atomic systems.
Hi Bluesky! I'd like to begin my journey in this platform by advertising this school I am currently co-organizing:
The Winter School on Ultracold Quantum Many-body Systems ❄️⚛️ (16-22 Feb 2025) is open for registration! The school will take place at the Centro de Ciencias de Benasque Pedro Pascual🏔️.
We are organizing a workshop on ML for quantum matter in Dresden in February 2025. The application deadline is Nov. 30, apply! www.pks.mpg.de/mlqmat25
Subsystem Evolution Speed as Indicator of Relaxation
New method to probe relaxation in quantum matter. No need to have a priori knowledge about relaxed steady state, just need to calculate how fast reduced density matrices evolve in time:
doi.org/10.48550/arX...
Thanks!
Thanks! Here we are indeed studying just expectation values of local operators, which might not necessarily mean that the quantum state at a more general level is also accurate. We've addressed this in a follow-up here:
arxiv.org/abs/2406.06198
So it seems that we get the full wave function right.
Adaptive Trotterization for Time-Dependent Hamiltonian Quantum Dynamics
We introduce an adaptive Trotterization algorithm for simulating time-dependent Hamiltonians on quantum computers, using piecewise conserved quantities to control & estimate errors. Read our PRL here:
doi.org/10.1103/Phys...
Empowering deep neural quantum states through efficient optimization
Deep learning for quantum matter. Our new algorithm enables the training of massive neural networks in order to solve the quantum many-body problem with unprecedented accuracy.
doi.org/10.1038/s415...
Wave-Function Network Description and Kolmogorov Complexity of Quantum Many-Body Systems
Wave-function networks can become scale-free as we show for theoretical data obtained through neural quantum states and experimental data from a Rydberg quantum simulator:
journals.aps.org/prx/abstract...
Vortex loop dynamics and dynamical quantum phase transitions
During nonequilibrium dynamics vortices can form in the phase of the Green's function. In this work we show that these vortices form topological structures in the form of loops:
doi.org/10.1103/Phys...
The fission of a string is an astounding phenomenon in confining gauge theories. It has been observed that the dynamics of such string breaking can either be fast or very slow. We put forward a dynamical localization transition as the underlying mechanism:
journals.aps.org/prl/abstract...
