arXiv subjects

Achim Kempf

Publications and source records attributed to Achim Kempf.

At least 19 recordsLinked to original sources

arXiv2026

The Einstein Tensor from UV-regulated Field Correlations

It is known that the spacetime metric can be recovered from quantum-field correlators. We extend this programme by showing that the Einstein tensor can be directly written in terms of UV-regulated field correlations. The same correlation structure also directly determines the Ricci tensor, while higher coincidence derivatives recover the connection and Riemann tensor. We further show that the renormalized stress-energy tensor can be formulated within this framework, up to the standard finite renormalization freedom. These results provide a correlation-based organization of the semiclassical Einstein equations and advance the programme of formulating spacetime geometry in terms of quantum field correlations.

gr-qc↗

arXiv2026

Principles of Quantum Optimization for Constrained Problems

Constrained combinatorial optimization underlies many industrial and technological decision problems. We develop a spectral theory that unifies many quantum optimization algorithms. We show that computational slowdown is driven by entanglement restructuring: the creation, redistribution, and destruction of entanglement during system evolution. The severity of the slowdown depends on how much entanglement must be changed. We show that algebraic properties of constraints induce such restructuring, and that constraint-aware dynamics reduce the associated slowdown by avoiding unnecessary restructuring. This framework explains why constraint-aware quantum methods can outperform generic penalty-based approaches. The theory connects constrained optimization, computational complexity, entanglement dynamics, and Hamiltonian spectral structure across continuous-time and circuit-based quantum optimization paradigms.

quant-ph↗

arXiv2026

Experimental demonstration that qubits can be cloned at will, if encrypted with a single-use decryption key

The no-cloning theorem forbids the creation of identical copies of qubits, thereby imposing strong limitations on quantum technologies. A recently-proposed protocol, encrypted cloning, showed, however, that the creation of perfect clones is theoretically possible - if the clones are simultaneously encrypted with a single-use decryption key. It has remained an open question, however, whether encrypted cloning is stable under hardware noise and thus practical as a quantum primitive. This is nontrivial because spreading quantum information widely could dilute it until barely exceeding the noise level, leading to catastrophic fidelity decay. Given the complexity of hardware noise, theory and classical simulation are insufficient to settle this. Here, we settle this question experimentally, on IBM Heron-R2 superconducting processors using up to 154 qubits. We find that encrypted cloning is stable under hardware noise, even when used as a module, namely in parallel, series or interleaved, while preserving pre-existing entanglement. This establishes it as a versatile quantum primitive for practical use, and it necessitates a refinement to our understanding of the no-cloning theorem: quantum information can be spread at will, in theory and in practice, without dilution or degradation, if encrypted or obscured. The actual constraint is that the decryption mechanism must be single-use.

quant-ph↗

arXiv2026

Trapped by simplicity: When Transformers fail to learn from noisy features

Noise is ubiquitous in data used to train large language models, but it is not well understood whether these models are able to correctly generalize to inputs generated without noise. Here, we study noise-robust learning: are transformers trained on data with noisy features able to find a target function that correctly predicts labels for noiseless features? We show that transformers succeed at noise-robust learning for a selection of $k$-sparse parity and majority functions, compared to LSTMs which fail at this task for even modest feature noise. However, we find that transformers typically fail at noise-robust learning of random $k$-juntas, especially when the boolean sensitivity of the optimal solution is smaller than that of the target function. We argue that this failure is due to a combination of two factors: transformers' bias toward simpler functions, combined with an observation that the optimal function for noise-robust learning typically has lower sensitivity than the target function for random boolean functions. We test this hypothesis by exploiting transformers' simplicity bias to trap them in an incorrect solution, but show that transformers can escape this trap by training with an additional loss term penalizing high-sensitivity solutions. Overall, we find that transformers are particularly ineffective for learning boolean functions in the presence of feature noise.

cs.LG↗

arXiv2025

The metric from energy-momentum non-conservation: Generalizing Noether and completing spectral geometry

We complete the program of spectral geometry, in the sense that we show that a manifold's shape, i.e., its metric, can be reconstructed from its resonant sound when tapped lightly, i.e., from its spectrum, -- if in addition we also record the resonances' mutual excitation pattern when the driving is strong enough to reach the nonlinear regime. Applied to spacetime, this finding yields a generalization of Noether's theorem: the specific pattern of energy-momentum non-conservation on a generic curved spacetime, encoded within the quantum field theoretic scattering matrices, is sufficient to calculate the metric. Applied to quantum gravity, this shows that the conventional dichotomy of spacetime versus matter can emerge from an underlying information-theoretic framework of only one type of quantity: abstract correlators, $G^{(n)}$, that are, a priori, merely operators on $n$ tensor factors of a Hilbert space. This is because, on one hand, if abstract higher-order correlators $G^{(n>2)}$ can be diagonalized, these correlators can be represented as local quantum field theoretic vertices on a curved spacetime whose metric $g_{\mu\nu}(x)$ can be explicitly calculated. On the other hand, at sufficiently high energies, such as the Planck scale, the $G^{(n)}$ may not be even approximately representable as correlators of a local QFT on a spacetime, indicating a regime that is mathematically controlled but transcends the concepts of spacetime and matter.

gr-qc↗

arXiv2025

Emergent metric from wavelet-transformed quantum field theory

We introduce a method of reverse holography by which a bulk metric is shown to arise from locally computable multiscale correlations of a boundary quantum field theory (QFT). The metric is obtained from the Petz-R\'enyi mutual information defined with input correlations computed from the continuous wavelet transform. The method is applicable to a variety of boundary QFTs that need not be conformal field theories (CFTs). For thermal free scalar and Dirac field theories the resulting bulk metric is that of (asymmetrically) warped anti-de Sitter (AdS) space. For massless, ground state CFTs the geometry simply reduces to AdS space. We show that certain parameters of the geometry can be tuned by changing the shape of the wavelet function.

hep-th↗

arXiv2025

Encrypted Qubits can be Cloned

We show that encrypted cloning of unknown quantum states is possible. Any number of encrypted clones of a qubit can be created through a unitary transformation, and each of the encrypted clones can be decrypted through a unitary transformation. The decryption of an encrypted clone consumes the decryption key, i.e., only one decryption is possible, in agreement with the no-cloning theorem. Encrypted cloning represents a new paradigm that provides a form of redundancy, parallelism or scalability where direct duplication is forbidden by the no-cloning theorem. For example, a possible application of encrypted cloning is to enable encrypted quantum multi-cloud storage.

quant-ph↗

arXiv2024

Reconstructing Superoscillations Buried Deeply in Noise

We utilize a method using frequency combs to construct waves that feature superoscillations - local regions of the wave that exhibit a change in phase that the bandlimits of the wave should not otherwise allow. This method has been shown to create superoscillating regions that mimic any analytic function - even ones well outside the bandlimits - to an arbitrary degree of accuracy. We experimentally demonstrate that these waves are extremely robust against noise, allowing for accurate reconstruction of a superoscillating target function thoroughly buried in noise. We additionally show that such a construction can be easily used to range-resolve a signal well below the commonly accepted fundamental limit.

physics.optics↗

arXiv2024

Separable ellipsoids around multipartite states

We show that, in finite dimensions, around any $m$-partite product state $\rho_{\rm prod}=\rho_1\otimes...\otimes\rho_m$, there exists an ellipsoid of separable states centered around $\rho_{\rm prod}$. This separable ellipsoid contains the separable ball proposed in previous works, and the volume of the ellipsoid is typically exponentially larger than that of the ball, due to the hierarchy of eigenvalues in typical states. We further generalize this ellipsoidal criterion to a trace formula that yields separable region around all separable states, and further study biseparability. Our criteria not only help numerical procedures to rigorously detect separability, but they also lead to a nested hierarchy of SLOCC-stable subsets that cover the separable set. We apply the procedure for separability detection to 3-qubit X states, genuinely entangled 4-qubit states mixed with noise, and the 1d transverse field Ising model at finite temperature to illustrate the power of our procedure for understanding entanglement in physical systems.

quant-ph↗

arXiv2024

Adiabatic Dynamics of Entanglement

We show that, during adiabatic evolution, any changes in entanglement can be attributed to a succession of avoided energy level crossings at which eigenvalues swap their eigenvectors. These swaps mediate the generation and redistribution of entanglement in multipartite systems. The efficiency of this redistribution depends on the narrowness of the avoided level crossings and thus constrains the speed of adiabatic evolution. Moreover, we relate the amount of entanglement involved to the ruggedness of the energy landscape, which directly affects the hardness of a computational problem. This enables an analysis of computational complexity and quantum advantage from the point of view of entanglement requirements. Applied to adiabatic quantum computation, our findings directly relate the computation's speed to its utilization of entanglement as a resource. The same principles extend to gate-based discretized adiabatic quantum algorithms, including those for Hamiltonian simulation and combinatorial optimization.

quant-ph↗

arXiv2024

The Best Radar Ranging Pulse to Resolve Two Reflectors

Previous work established fundamental bounds on subwavelength resolution for the radar range resolution problem, called superradar [Phys. Rev. Appl. 20, 064046 (2023)]. In this work, we identify the optimal waveforms for distinguishing the range resolution between two reflectors of identical strength. We discuss both the unnormalized optimal waveform as well as the best square-integrable pulse, and their variants. Using orthogonal function theory, we give an explicit algorithm to optimize the wave pulse in finite time to have the best performance. We also explore range resolution estimation with unnormalized waveforms with multi-parameter methods to also independently estimate loss and time of arrival. These results are consistent with the earlier single parameter approach of range resolution only and give deeper insight into the ranging estimation problem. Experimental results are presented using radio pulse reflections inside coaxial cables, showing robust range resolution smaller than a tenth of the inverse bandedge, with uncertainties close to the derived Cram\'er-Rao bound.

eess.SP↗

arXiv2023

Separable Ball around any Full-Rank Multipartite Product State

We show that around any $m$-partite product state $\rho_{\rm prod}=\rho_1\otimes...\otimes\rho_m$ of full rank (that is ${\rm det}(\rho_{\rm prod})\neq 0)$, there exists a finite-sized closed ball of separable states centered around $\rho_{\rm prod}$ whose radius is $\beta:=2^{1-m/2}\lambda_{\rm min}(\rho_{\rm prod})$. Here, $\lambda_{\rm min}(\rho_{\rm prod})$ is the smallest eigenvalue of $\rho_{\rm prod}$. We are assuming that the total Hilbert space is finite dimensional and we use the notion of distance induced by the Frobenius norm. Applying a scaling relation, we also give a new and simple sufficient criterion for multipartite separability based on trace: ${\rm Tr}[\rho\rho_{\rm prod}]^2/{\rm Tr}[\rho^2]\geq {\rm Tr}[\rho_{\rm prod}^2]-\beta^2$. Using the separable balls around the full-rank product states, we discuss the existence and possible sizes of separable balls around any multipartite separable states, which are important features for the set of all separable states. We discuss the implication of these separable balls on entanglement dynamics.

quant-ph↗

arXiv2023

Quantum lattice models that preserve continuous translation symmetry

Bandlimited approaches to quantum field theory offer the tantalizing possibility of working with fields that are simultaneously both continuous and discrete via the Shannon Sampling Theorem from signal processing. Conflicting assumptions in general relativity and quantum field theory motivate the use of such an appealing analytical tool that could thread the needle to meet both requirements. Bandlimited continuous quantum fields are isomorphic to lattice theories, yet without requiring a fixed lattice. Any lattice with a required minimum spacing can be used. This is an isomorphism that avoids taking the limit of the lattice spacing going to zero. In this work, we explore the consequences of this isomorphism, including the emergence of effectively continuous symmetries in quantum lattice theories. One obtains conserved lattice observables for these continuous symmetries, as well as a duality of locality from the two perspectives. We expect this work and its extensions to provide useful tools for considering numerical lattice models of continuous quantum fields arising from the availability of discreteness without a fixed lattice, as well as offering new insights into emergent continuous symmetries in lattice models and possible laboratory demonstrations of these phenomena.

quant-ph↗

arXiv2023

Emergence of Spacetime from Fluctuations

We use a result of Hawking and Gilkey to define a Euclidean path integral of gravity and matter which has the special property of being independent of the choice of basis in the space of fields. This property allows the path integral to describe also physical regimes that do not admit position bases. These physical regimes are pre-geometric in the sense that they do not admit a mathematical representation of the physical degrees of freedom in terms of fields that live on a spacetime. In regimes in which a spacetime representation does emerge, the geometric properties of the emergent spacetime, such as its dimension and volume, depend on the balance of fermionic pressure and bosonic and gravitational pull. That balance depends, at any given energy scale, on the number of bosonic and fermionic species that contribute, which in turn depends on their masses. This yields an explicit mechanism by which the effective spacetime dimension can depend on the energy scale.

gr-qc↗

arXiv2023

Lensing of Vacuum Entanglement near Schwarzschild Black Holes

An important feature of Schwarzschild spacetime is the presence of orbiting null geodesics and caustics. Their presence implies strong gravitational lensing effects for matter and radiation, i.e., for excitations of quantum fields. Here, we raise the question whether the lensing manifests itself also in the vacuum of quantum fields, namely by lensing the distribution of vacuum entanglement. To explore this possibility, we use the method of entanglement harvesting, where initially unentangled localized quantum systems are temporarily coupled to the field at different locations. We find that for the Boulware, Hartle-Hawking and Unruh vacua in 3+1 dimensional Schwarzschild spacetime, the harvesting of vacuum entanglement is indeed greatly amplified near caustics. In particular, we establish that pre-existing vacuum entanglement can be harvested also for lightlike separations.

quant-ph↗

arXiv2023

Entanglement is better teleported than transmitted

We show that, for the purpose of quantum communication via a quantum field, it is essential to view the field not only as a medium for transmission but also as a source of entanglement that can aid in the communication task. To this end, we consider the quantum communication scenario where Alice is initially entangled with an ancilla and intends to communicate with Bob through a quantum field, so as to make Bob entangled with the ancilla. We find that if Alice and Bob communicate by directly coupling to the quantum field, then they can generate negativity between Bob and the ancilla only at orders that are higher than second perturbative order. We then present a protocol based on quantum teleportation in which Alice and Bob consume entanglement that they obtained from the field via interaction or harvesting. We show that this protocol can transfer negativity already to second perturbative order.

quant-ph↗

arXiv2022

The Transfer of Entanglement Negativity at the Onset of Interactions

Quantum information, in the form of entanglement with an ancilla, can be transmitted to a third system through interaction. Here, we investigate this process of entanglement transmission perturbatively in time. Using the entanglement monotone negativity, we determine how the proclivity of an interaction to either generate, transfer or lose entanglement depends on the choice of Hamiltonians and initial states. These three proclivities are captured by Hamiltonian- and state-dependent quantities that we call negativity susceptibility, negativity transmissibility and negativity vulnerability respectively. These notions could serve, for example, as cost functions in quantum technologies such as machine-learned quantum error correction.

quant-ph↗

arXiv2022

Covariant Predictions for Planck-Scale Features in Primordial Power Spectra

In this companion to our letter (arXiv:2208.10514), we elaborate the full details of the predicted corrections to the primordial scalar and tensor power spectra that arise from quantum gravity-motivated, natural, covariant ultraviolet cutoffs. We implement these cutoffs by covariantly restricting the fields which are summed over in the path integrals for the primordial correlators, and we discuss in detail the functional analytic techniques necessary for evaluating such path integrals. Our prediction, which is given in terms of measured cosmological parameters and without assuming any particular inflationary potential, is that the corrections take the form of small oscillations which are superimposed on the conventional power spectra. The frequency of these oscillations only depends on the location of the cutoff scale and the first inflationary slow-roll parameter, while the amplitude and phase are also moderately sensitive to how smoothly the cutoff turns on. The specificity of the new predictions offers an opportunity to significantly enhance experimental sensitivity in observations of the cosmic microwave background and large-scale structure. This may be used to place ever higher bounds on the scale at which quantum gravity effects become important in quantum field theory or may even provide positive evidence for quantum gravity effects.

hep-th↗