arXiv:2501. 08640v2 Announce Type: replace Abstract: We propose a way to bound the generalisation errors of several classes of quantum reservoirs using the Rademacher complexity.
arXiv:2607. 18552v1 Announce Type: cross Abstract: Quantum reservoir computing (QRC) uses the dynamics of a fixed or weakly tuned quantum system to transform temporal and sequential inputs into measured features, while training is typically confined to a classical readout.
arXiv:2607. 07978v1 Announce Type: cross Abstract: Quantum reservoir computing uses a fixed quantum circuit as a feature generator and trains only a simple linear readout on top of it.
The paper investigates whether having only forward access to a state-preparation unitary—without its inverse—can reduce the number of queries needed for quantum PAC learning. By analyzing worst-case scenarios over all compatible unitaries and finite dimensions, the authors prove that the optimal forward-only query complexities for realizable and agnostic learning are θ((d+log(1/δ))/ε) and θ((d+log(1/δ))/ε²), respectively, matching classical and quantum-copy bounds. These results demonstrate that forward-only access offers no asymptotic advantage over classical data or quantum copies, highlighting the essential role of inverse access for any improvement in the realizable setting.
The paper investigates quantum reservoir computing using highly scrambling quantum systems modeled by high‑order unitary designs. It shows that in noiseless settings, measurement readouts become exponentially concentrated as the reservoir size grows, but this concentration does not worsen with repeated iterations, while memory of early inputs decays exponentially with both reservoir size and iterations. In noisy environments, memory also decays exponentially over time for local noisy channels, and the study introduces new proof techniques for bounding concentration in temporal quantum models. Numerical results indicate that even physical reservoirs, such as an Ising model in a quantum‑chaotic phase, can exhibit exponential concentration, whereas reservoirs in a many‑body localized phase or at the edge of chaos do not.
Quantum reservoir computing uses a fixed quantum circuit as a feature generator and trains only a simple linear readout on top of it. This makes it cheap to train and free of the optimisation problems that affect many quantum machine-learning models.
arXiv:2602. 01177v3 Announce Type: replace-cross Abstract: We develop an information-theoretic framework connecting stability, privacy, and generalization for quantum learning algorithms.
arXiv:2607. 21409v1 Announce Type: cross Abstract: A central challenge in quantum machine learning is understanding the scaling behavior of parameterized quantum circuits (PQCs).
We study stabilizer state testing and learning with limited coherent quantum memory. Here an algorithm sequentially receives copies of an unknown $n$-qubit state, but may keep only $k$ qubits of coherent quantum memory between measurements.
arXiv:2601.22005v2 Announce Type: replace-cross Abstract: Distance metrics are central to machine learning, yet distances between ensembles of quantum states remain poorly understood due to fundament...
arXiv:2606. 12211v1 Announce Type: cross Abstract: A central principle in quantum machine learning is that an ansatz should be expressive enough to represent the quantum data of interest.
arXiv:2607. 02444v1 Announce Type: cross Abstract: We study stabilizer state testing and learning with limited coherent quantum memory.
arXiv:2607. 09905v1 Announce Type: cross Abstract: Can a small quantum computer forecast a changing signal better than an ordinary classical method?