I'm Markus Baumann, a researcher at LMU Munich and the QAR-Lab.
My research is in quantum computing. I study quantum machine learning and quantum reservoir computing: how quantum systems learn, remember and process information over time.
Beyond these topics, I work on AI for scientific research itself. I use and study AI across the whole research process, from forming hypotheses and designing and running experiments to weighing the evidence and writing up results that others can reproduce. My goal is to find out how far AI can take science, and to make research faster, more transparent and easier to verify.
Website · Google Scholar · LMU profile · QAR-Lab · Email
- Quantum reservoir computing: how quantum dynamics store and process information over time.
- Quantum machine learning: quantum models, how they learn and how well they generalize.
- Time series: forecasting, temporal structure, stochastic models and dynamical systems.
- Representation learning: features that help a model predict and generalize.
- AI for science: AI-assisted and agentic workflows for reproducible research.
Papers, preprints, and reproducibility packages.
| Paper | Status | Focus | Links |
|---|---|---|---|
| Parity Supervision as a Driver of Generalization in Quantum Generative Modeling (2026) | To appear in the IEEE QCE 2026 proceedings — Quantum–GenAI Co-Design & Co-Discovery (QGDD) Technical Papers 🏆 Best Paper Award — 2nd place, QGDD track |
Parity supervision and generalization to unseen valid states in quantum generative models. | arXiv · repository (public) |
| Symmetry Alone Is Not an Ansatz: Task-Aligned Interactions in Equivariant Quantum Circuits (2026) | To appear in the IEEE QCE 2026 workshop proceedings — 4th International Workshop on Quantum Machine Learning: From Research to Practice | How task-aligned trainable interactions improve learning within symmetry-preserving circuit families. | arXiv · repository (public) |
| Where a Quantum Reservoir Works: A Transferable Operating Band (2026) | To appear in the IEEE QCE 2026 workshop proceedings — 4th International Workshop on Quantum Machine Learning: From Research to Practice | A transferable operating band for quantum reservoirs, located using a task-free memory diagnostic. | arXiv · repository (public) |
| From Quantum Shots to Training Data: Reorganizing Measurement Records in Quantum Machine Learning (2026) | Accepted at IEEE QAI 2026 | Improving finite-shot learning by regrouping existing measurement records at unchanged quantum execution cost. | arXiv · repository (public) |
| Exploiting Symmetry in Quantum Reservoir Computing (2026) | Accepted at IEEE QAI 2026 | End-to-end symmetry-preserving QRC, evaluated in noisy simulation, on IBM hardware, and on real weather data. | arXiv · repository (public) |
| When Can Quantum Extreme Learning Machines Replace Quantum Reservoirs? (2026) | Accepted at ICAART 2027 | When recent inputs suffice to reproduce a quantum reservoir's outputs, separating missing history, model capacity, and measurement error. | arXiv · repository (private) |
| Paper | Status | Focus | Links |
|---|---|---|---|
| Can One Hear the Size of a Clock? (2026) | Submitted to Physical Review Letters | A counting law for stochastic clocks, the counterpart of Weyl's law: a network of d states sustains fewer than 2(d−1)/(πQ) rhythms sharper than Q, and measured rhythms reveal how large a hidden clock must be. | arXiv · repository (public) |
| The Fundamental Memory Ceiling of Noisy Quantum Reservoirs (2026) | Submitted to Physical Review Letters | Size-independent bounds on input memory under local depolarizing noise, with fresh-qubit supply as the limiting resource. | arXiv · paper + Supplemental Material · repository (public) |
| The Form of Dissipation Shapes What a Quantum Reservoir Remembers (2026) | Under review at Quantum | How the form of dissipation decides the way a quantum reservoir stores, writes and reads its memory, with two theorems on the trade-off between memory and nonlinearity. | arXiv · repository |
| Project | Focus |
|---|---|
| Vibe Shuffle | Browser-based study comparing sensor-guided and random song selection using listeners' ratings of mood fit. |
| Longevity Logbook | Terminal-first training logbook with editable strength programming. |
| 100x100 Pomodoro | Minimal Pomodoro tracker for repeated focused work sessions. |
| Cognitive Decision Tree | Animated decision-tree MVP for low-friction cognitive-state self-assessment. |
| Project | Focus |
|---|---|
| Research Website | Public website and research identity hub. |
| GitHub Pages site | Live deployment target for the public research website. |
| QRC Learning Platform | Learning site for Quantum Reservoir Computing concepts and flashcards. |
| Spaced-Repetition Tutor | Spaced-repetition and LLM tutoring loops for learning workflows. |
| macOS Writing Assistant | Sentence-level German and English correction with fast undo. |



