Search

Machine learning potentials have revolutionised the field of atomistic simulations in recent years and are becoming a mainstay in the toolbox of computational scientists. This paper aims to provide an overview and introduction into machine learning potentials and their practical application to scientific problems. We provide a systematic guide for developing machine learning potentials, reviewing chemical descriptors, regression models, data generation and validation approaches. We begin with an emphasis on the earlier generation of models, such as high-dimensional neural network potentials (HD-NNPs) and Gaussian approximation potential (GAP), to provide historical perspective and guide the reader towards the understanding of recent developments, which are discussed in detail thereafter. Furthermore, we refer to relevant expert reviews, open-source software, and practical examples - further lowering the barrier to exploring these methods. The paper ends with selected showcase examples, highlighting the capabilities of machine learning potentials and how they can be applied to push the boundaries in atomistic simulations.

Typestate systems are notoriously complex as they require sophisticated machinery for tracking aliasing. We propose a new, transition-oriented foundation for typestate in the setting of impure functional programming. Our approach relies on ordered types for simple alias tracking and its formalization draws on work on bunched implications. Yet, we support a flexible notion of borrowing in the presence of typestate. Our core calculus comes with a notion of resource types indexed by an ordered partial monoid that models abstract state transitions. We prove syntactic type soundness with respect to a resource-instrumented semantics. We give an algorithmic version of our type system and prove its soundness. Algorithmic typing facilitates a simple surface language that does not expose tedious details of ordered types. We implemented a typechecker for the surface language along with an interpreter for the core language.

Water confined in nanoscale cavities plays a crucial role in everyday phenomena in geology and biology, as well as technological applications at the water-energy nexus. However, even understanding the basic properties of nano-confined water is extremely challenging for theory, simulations, and experiments. In particular, determining the melting temperature of quasi-one-dimensional ice polymorphs confined in carbon nanotubes has proven to be an exceptionally difficult task, with previous experimental and classical simulations approaches report values ranging from $\sim 180 \text{ K}$ up to $\sim 450 \text{ K}$ at ambient pressure. In this work, we use a machine learning potential that delivers first principles accuracy to study the phase diagram of water for confinement diameters $ 9.5 < d < 12.5 \text{ \AA}$. We find that several distinct ice polymorphs melt in a surprisingly narrow range between $\sim 280 \text{ K}$ and $\sim 310 \text{ K}$, with a melting mechanism that depends on the nanotube diameter. These results shed new light on the melting of ice in one-dimension and have implications for the operating conditions of carbon-based filtration and desalination devices.

Many of graphene's remarkable properties are intrinsically linked to its inherent ripples. Defects, whether naturally present or artificially introduced, are known to have a strong impact on the rippling of graphene. However, how defects alter ripple dynamics in two-dimensional (2D) materials in general, and graphene in particular, remains largely unexplored. Here, using machine learning-driven molecular dynamics simulations, we reveal a fundamental connection between defect concentration and ripple dynamics in freestanding graphene sheets. Specifically, we find that at a critical concentration of approximately 0.1%, dynamic rippling undergoes a transition from freely propagating to static ripples. This is in quantitative alignment with the experimentally observed turning point in the non-monotonic scaling of the Young's modulus and emphasises the critical interplay between defects and material dynamics. Our work not only unveils the significant impact of defects on rippling dynamics in graphene but also paves the way to design two-dimensional devices with tailored properties., Comment: 8 pages, 20 pages supplementary information

Perturbative quantum gravity starts from prescribing a background metric. That background metric is then used in order to carry out two separate steps: 1. One splits the non-perturbative metric into background and deviation from it (graviton) and expands the action in terms of the graviton which results in an ifinite series of unknown radius of convergence. 2. One constructs a Fock representation for the graviton and performs perturbative graviton quantum field theory on the fixed background as dictated by the perturbative action. The result is a non-renormalisable theory without predictive power. It is therefore widely believed that a non-perturbative approach is mandatory in order to construct a fundamental, not only effective, predictive quantum field theory of the gravitational interaction. Since perturbation theory is by definition background dependent, the notions of background dependence (BD) and perturbation theory (PT) are often considered as symbiotic, as if they imply each other. In the present work we point out that there is no such symbiosis, these two notions are in fact logically independent. In particular, one can use BD structures while while not using PT at all. Specifically, we construct BD Fock representations (step 2 above) for the full, non-perturbative metric rather than the graviton (not step 1 above) and therefore never perform a perturbative expansion. Despite the fact that the gravitational Lagrangean is a non-polynomial, not even analytic, function of the metric we show that e.g. the Hamiltonian constraint with any density weight can be defined as a quadratic form with dense form domain in such a representation.

The canonical quantisation of General Relativity including matter on a spacetime manifold in the globally hyperbolic setting involves in particular the representation theory of the spatial diffeomorphism group (SDG), and/or its Lie algebra (SDA), of the underlying spatial submanifold. There are well known Fock representations of the SDA in one spatial dimension and non-Fock representations of the SDG in all dimensions. The latter are not strongly continuous and do not descend to representations of the SDA. In this work we report some partial results on non anomalous representations of the SDA for both geometry and matter: 1. Background independent Fock representations of the SDA by operators exist in all dimensions. 2. Infinitely many unitary equivalence classes of background dependent Fock representations of the SDA by operators exist in one dimension but these do not extend to higher dimensions. 3. Infinitely many unitary equivalence classes of background dependent Fock representations of the SDA of volume preserving diffeomorphisms by operators exist in all dimensions. 4. Infinitely many unitary equivalence classes of background dependent Fock representations of the SDA by quadratic forms exist in all dimensions. Except for 1. these representations do not descend from an invariant state of the Weyl algebra and 4. points to a new strategy for solving the quantum constraints.

In this paper we extend reduced phase space approach to black hole perturbation theory to Maxwell matter. We expand the resulting reduced Hamiltonian to second order in the graviton and photon perturbations and find that the corresponding equations of motion match the ones derived in the literature. Accordingly our approach reproduces previous results at second order. Its real virtue lies in the fact that it extends to any order in perturbation theory in a manifestly gauge invariant fashion.

In a seminal work, Hawking showed that natural states for free quantum matter fields on classical spacetimes that solve the spherically symmetric vacuum Einstein equations are KMS states of non-vanishing temperature. Although Hawking's calculation does not include backreaction of matter on geometry, it is more than plausible that the corresponding Hawking radiation leads to black hole evaporation which is in principle observable. Obviously, an improvement of Hawking's calculation including backreaction is a problem of quantum gravity. Since no commonly accepted quantum field theory of general relativity is available yet, it has been difficult to reliably derive the backreaction effect. An obvious approach is to use black hole perturbation theory of a Schwarzschild black hole of fixed mass and to quantise those perturbations. But it is not clear how to reconcile perturbation theory with gauge invariance beyond linear perturbations. In a recent work we proposed a new approach to this problem that applies when the physical situation has an approximate symmetry, such as homogeneity (cosmology), spherical symmetry (Schwarzschild) or axial symmetry (Kerr). The idea, which is surprisingly feasible, is to first construct the non-perturbative physical (reduced) Hamiltonian of the reduced phase space of fully gauge invariant observables and only then to apply perturbation theory directly in terms of observables. The task to construct observables is then disentangled from perturbation theory, thus allowing to unambiguosly develop perturbation theory to arbitrary orders. In this first paper of the the series we outline and showcase this approach for spherical symmetry and second order in the perturbations for Einstein-Klein-Gordon-Maxwell theory. Details and generalisation to other matter and symmetry and higher orders will appear in subsequent companion papers.

In this second paper of our series we focus on the classical pure gravity sector of spherically symmetric black hole perturbations and expand the reduced Hamiltonian to second order. To compare our manifestly gauge invariant formalism with established results in the literature we have to translate our results derived in Gullstrand-Painlev\'e gauge to the gauges used in those works. After several canonical transformations we expectedly find exact agreement with the Hamiltonian given by Moncrief which generates the Regge-Wheeler and Zerilli equations of motion for the linear axial (often denoted odd) and polar (often denoted even) perturbations respectively. This confirms the validity of our method which immediately generalises to higher orders.

For interacting classical field theories such as general relativity exact solutions typically can only be found by imposing physically motivated (Killing) {\it symmetry} assumptions. Such highly symmetric solutions are then often used as {\it backgrounds} in a {\it perturbative} approach to more general non-symmetric solutions. If the theory is in addition a {\it gauge} theory such as general relativity, the issue arises how to consistently combine the perturbative expansion with the gauge reduction. For instance it is not granted that the corresponding constraints expanded to a given order still close under Poisson brackets with respect to the non-symmetric degrees of freedom up to higher order. If one is interested in the problem of {\it backreaction} between symmetric and non-symmetric dgrees of freedom, then one also must consider the symmetric degrees of freedom as dynamical variables which supply additional terms in Poisson brackets with respect to the symmetric degrees of freedom and the just mentioned consistency problem becomes even more complicated. In this paper we show for a general theory how to consistently combine all of these notions. The idea is to {\it first} perform the {\it exact} gauge reduction on the {\it full} phase space which results in the reduced phase space of observables and physical Hamiltonian respectively and {\it secondly} expand that physical Hamiltonian perturbatively. Surprisingly, this strategy is not only practically feasible but also avoids the above mentioned tensions. We also show how to perform the partial reduction with respect to only the asymmetric constraints but that theory is not quantisable at finite orders unless there is only one symmetric constraint.

In a recent contribution we identified possible points of contact between the asymptotically safe and canonical approach to quantum gravity. The idea is to start from the reduced phase space (often called relational) formulation of canonical quantum gravity which provides a reduced (or physical) Hamiltonian for the true (observable) degrees of freedom. The resulting reduced phase space is then canonically quantised and one can construct the generating functional of time ordered Wightman (i.e. Feynman) or Schwinger distributions respectively from the corresponding time translation unitary group or contraction semigroup respectively as a path integral. For the unitary choice that path integral can be rewritten in terms of the Lorentzian Einstein Hilbert action plus observable matter action and a ghost action. The ghost action depends on the Hilbert space representation chosen for the canonical quantisation and a reduction term that encodes the reduction of the full phase space to the phase space of observavbles. This path integral can then be treated with the methods of asymptically safe quantum gravity in its {\it Lorentzian} version. We also exemplified the procedure using a concrete, minimalistic example namely Einstein-Klein-Gordon theory with as many neutral and massless scalar fields as there are spacetime dimensions. However, no explicit calculations were performed. In this paper we fill in the missing steps. Particular care is needed due to the necessary switch to Lorentzian signature which has strong impact on the convergence of ``heat'' kernel time integrals in the heat kernel expansion of the trace involved in the Wetterich equation and which requires different cut-off functions than in the Euclidian version. As usual we truncate at relatively low order and derive and solve the resulting flow equations in that approximation.

The canonical (CQG) and asymptotically safe (ASQG) approach to quantum gravity share to be both non-perturbative programmes. However, apart from that they seem to differ in several aspects such as: 1. Signature: CQG is Lorentzian while ASQG is mostly Euclidian. 2. Background Independence (BI): CQG is manifesly BI while ASQG is apparently not. 3. Truncations: CQG is apparently free of truncations while ASQG makes heavy use of them. The purpose of the present work is to either overcome actual differences or to explain why apparent differences are actually absent. Thereby we intend to enhance the contact and communication between the two communities. The focus of this contribution is on conceptual issues rather than deep technical details such has high order truncations. On the other hand the paper tries to be self-contained in order to be useful to researchers from both communities. The point of contact is the path integral formulation of Lorentzian CQG in its reduced phase space formulation which yields the formal generating functional of physical (i.e. gauge invariant) either Schwinger or Feynman N-point functions for (relational) observables. The corresponding effective actions of these generating functionals can then be subjected to the ASQG Wetterich type flow equations which serve in particular to find the rigorous generating fuctionals via the inverse Legendre transform of the fixed pointed effective action.

We use 62 electron density profiles collected by the Radio Occultation Science Experiment (ROSE), on MAVEN, when Mars was hit by CIRs and ICMEs close to aphelion (April 2021) and during two dust storms (June-July 2022) to examine the response of the Martian ionosphere to solar events and to solar events hitting during dust storms. We do so through three proxies - variation in total electron content between 80 and 300 km altitude, peak density, and peak altitude - of the aforementioned 62 ROSE electron density profiles, relative to a characterisation of the ionosphere through solar minimum leading to solar maximum, specific to local time sector and season, presented in Segale et al., (COMPANION). We observe an increased Total Electron Content (TEC) between 80 and 300 km altitude up to 2.5 x 10(15) m(-2) in April 2021 and up to 5 x 10(15) m(-2) in June-July 2022 compared to the baseline photochemically produced ionosphere. This increase in TEC corresponds mainly to increases in the solar energetic particles flux (detected by MAVEN SEP) and electron fluxes (detected by MAVEN SWEA). In addition to solar events, in June-July 2022, an A storm and a B storm were occurring and merging on the surface of Mars. We observe a raise in peak altitude in general lower than expected during dust storms, possibly due to high values of solar wind dynamic pressure (derived from MAVEN SWIA). From 31 ROSE profiles collected in this time period that showed both the M2 and M1 layer, we observe that, on average, M1 and M2 peak altitudes raise the same amount, suggesting that the thermosphere might loft as a unit during dust storms. During this time period, several proton aurora events of variable brightness were detected with MAVEN IUVS underlining the complex and multifaceted impact of dust activity and extreme solar activity on the Martian ionosphere.

Over the past decade, Children's Savings Account (CSA) programs have proliferated across the United States. CSA programs pair long-term savings and investment accounts with other supports to help families pay for post-secondary education. The goal of most CSA programs is to increase participants'--especially those from low-income households--post-secondary enrollment and completion. Within that broader aim, many programs also have goals around addressing barriers for Black and Latine participants to accessing and completing post-secondary education. However, there has been only limited research into how CSA programs are supporting Black and Latine participants. To begin to address this research gap, "Abt Associates" and "Prosperity Now" conducted research in 2020-21 with four CSA program sites. The four case studies explore what the programs are doing to address challenges faced by Black and Latine students in engaging with CSAs and ultimately accessing and completing post-secondary education.

Elucidating the interplay between nuclear and electronic degrees of freedom that govern the complex dielectric behavior of materials under intense photoexcitation is essential for tailoring optical properties on demand. However, conventional transient reflectivity experiments have been unable to differentiate between real and imaginary components of the dielectric response, omitting crucial electron-lattice interactions. Utilizing thin film interference we unambiguously determined the photoinduced change in complex dielectric function in the Peierls semimetal bismuth and examined its dependence on the excitation density and nuclear motion of the A$_{1g}$ phonon. Our modeled transient reflectivity data reveals a progressive broadening and redshift of Lorentz oscillators with increasing excitation density and underscores the importance of both, electronic and nuclear coordinates in the renormalization of interband transitions., Comment: Manuscript (6 pages) plus supplemental material (6 pages)

Upon impulsive optical excitation of solid-state materials, the non-equilibrium flow of energy from the excited electronic system to the lattice degrees of freedom typically happens in a few picoseconds. Here we identified the surface of thin Bi films grown on Si(001) as an additional subsystem which is excited much slower on a 100 ps timescale that is caused by decoupling due to mismatched phonon dispersions relations of bulk and surface. Anharmonic coupling among the phonon systems provides pathways for excitations which exhibits a 1/T-dependence causing a speed-up of surface excitation at higher temperatures. A quantitative justification is provided by phonon Umklapp processes from lattice thermal conductivity of the Bi bulk. Three-temperature model simulations reveal a pronounced non-equilibrium situation up to nanoseconds: initially, the surface is colder than the bulk, that situation is then inverted during cooling and the surface feeds energy back into the bulk phonon system.

In a recent contribution we showed that there exists a smooth, dense domain for singular potential Schr\"odinger operators on the real line which is invariant under taking derivatives of arbitrary order and under multiplication by positive and negative integer powers of the coordinate. Moreover, inner products between basis elements of that domain were shown to be easily computable analytically. A task left open was to construct an orthonormal basis from elements of that domain by using Gram-Schmidt orthonormalisation. We perform that step in the present manuscript. We also consider the application of these methods to the positive real line for which one can no longer perform the integrals analytically but for which one can give tight analytical estimates., Comment: 5 p

Recently it was shown that it is undecidable whether a term rewrite system can be proved terminating by a polynomial interpretation in the natural numbers. In this paper we show that this is also the case when restricting the interpretations to linear polynomials, as is often done in tools, and when only considering single-rule rewrite systems. What is more, the new undecidability proof is simpler than the previous one. We further show that polynomial termination over the rationals/reals is undecidable., Comment: Presented at WST 2023

The Weighted Path Order of Yamada is a powerful technique for proving termination. It is also supported by CeTA, a certifier for checking untrusted termination proofs. To be more precise, CeTA contains a verified function that computes for two terms whether one of them is larger than the other for a given WPO, i.e., where all parameters of the WPO have been fixed. The problem of this verified function is its exponential runtime in the worst case. Therefore, in this work we develop a polynomial time implementation of WPO that is based on memoization. It also improves upon an earlier verified implementation of the Recursive Path Order: the RPO-implementation uses full terms as keys for the memory, a design which simplified the soundness proofs, but has some runtime overhead. In this work, keys are just numbers, so that the lookup in the memory is faster. Although trivial on paper, this change introduces some challenges for the verification task., Comment: Presented at WST 2023

The lock set method and the partial order method are two main approaches to guarantee that dynamic data race prediction remains efficient. There are many variations of these ideas. Common to all of them is the assumption that the events in a critical section belong to the same thread. We have evidence that critical sections in the wild do extend across thread boundaries even if the surrounding acquire and release events occur in the same thread. We introduce the novel concept of a cross-thread critical section to capture such situations, offer a theoretical comprehensive framework, and study their impact on state-of-the-art data race analyses. For the sound partial order relation WCP we can show that the soundness claim also applies to cross-thread critical sections. For DCtp the occurrence of cross-thread critical sections invalidates the soundness claim. For complete partial order relations such as WDP and PWR, cross-thread critical sections help to eliminate more false positives. The same (positive) impact applies to the lock set construction. Our experimental evaluation confirms that cross-thread critical sections arise in practice. For the complete relation PWR, we are able to reduce the number of false positives. The performance overhead incurred by tracking cross-thread critical sections slows down the analysis by 10%-20%, on average., Comment: Revised POPL'24 submission. 1. WCP is sound and show that WCP soundness proof can be adapted. 2. Cross-thread critical sections arise in practice, though the impact is not drastic. This in line with other works (like WCP) that advance the state of the art

In this paper we present a method for single-channel wind noise reduction using our previously proposed diffusion-based stochastic regeneration model combining predictive and generative modelling. We introduce a non-additive speech in noise model to account for the non-linear deformation of the membrane caused by the wind flow and possible clipping. We show that our stochastic regeneration model outperforms other neural-network-based wind noise reduction methods as well as purely predictive and generative models, on a dataset using simulated and real-recorded wind noise. We further show that the proposed method generalizes well by testing on an unseen dataset with real-recorded wind noise. Audio samples, data generation scripts and code for the proposed methods can be found online (https://uhh.de/inf-sp-storm-wind)., Comment: Accepted to VDE 15th ITG conference on Speech Communication

In the next decade, there is an opportunity for very high return on investment of relatively small budgets by elevating the priority of smallsat funding in heliophysics. We've learned in the past decade that these missions perform exceptionally well by traditional metrics, e.g., papers/year/\$M (Spence et al. 2022 -- arXiv:2206.02968). It is also well established that there is a "leaky pipeline" resulting in too little diversity in leadership positions (see the National Academies Report at https://www.nationalacademies.org/our-work/increasing-diversity-in-the-leadership-of-competed-space-missions). Prioritizing smallsat funding would significantly increase the number of opportunities for new leaders to learn -- a crucial patch for the pipeline and an essential phase of career development. At present, however, there are far more proposers than the available funding can support, leading to selection ratios that can be as low as 6% -- in the bottom 0.5th percentile of selection ratios across the history of ROSES. Prioritizing SmallSat funding and substantially increasing that selection ratio are the fundamental recommendations being made by this white paper., Comment: White paper submitted to the Decadal Survey for Solar and Space Physics (Heliophysics) 2024-2033; 6 pages, 1 figure

The transport of sputtered species from the target of a magnetron plasma to a collecting surface at the circumference of the plasma is analyzed using a particle tracer technique. A small chromium insert at the racetrack position inside a titanium target is used as the source of tracer particles, which are redeposited on the collecting surface. The azimuthal velocity of the ions along the racetrack above the target is determined from the Doppler shift of the optical emission lines of titanium and chromium. The trajectories are reconstructed from an analysis of the transport physics leading to the measured deposition profiles. It is shown that a simple direct-line-of sight re-deposition model can explain the data for low power plasmas (DCMS) and for pulsed high power impulse magnetron plasmas (HiPIMS) by using the Thompson velocity distribution from the sputter process as starting condition. In the case of a HiPIMS plasma, the drag force exerted on the ions and neutrals by the electron Hall current has to be included causing an azimuthal displacement in \ExB direction. Nevertheless, the Thompson sputter distribution remains preserved for 50\% of the re-deposited growth flux. The implications for the understanding of transport processes in magnetron plasmas are discussed.

Magnetron sputtering discharges feature complex magnetic field configurations to confine the electrons close to the cathode surface. This magnetic field configuration gives rise to a strong electron drift in azimuthal direction, with typical drift velocities on the order of \SI{100}{\kilo\meter\per\second}. In high power impulse magnetron sputtering (HiPIMS) plasmas, the ions have also been observed to follow the movement of electrons with velocities of a few \si{\kilo\meter\per\second}, despite being unmagnetized. In this work, we report on measurements of the azimuthal ion velocity using spatially resolved optical emission spectroscopy, allowing for a more direct measurement compared to experiments performed using mass spectrometry. The azimuthal ion velocities increase with target distance, peaking at about \SI{1.55}{\kilo\meter\per\second} for argon ions and \SI{1.25}{\kilo\meter\per\second} for titanium ions. Titanium neutrals are also found to follow the azimuthal ion movement which is explained with resonant charge exchange collisions. The experiments are then compared to a simple test-particle simulation of the titanium ion movement, yielding good agreement to the experiments when only considering the momentum transfer from electrons to ions via Coulomb collisions as the only source of acceleration in azimuthal direction. Based on these results, we propose this momentum transfer as the primary source for ion acceleration in azimuthal direction.

Vielbeins are necessary when coupling General Relativity (GR) to fermionic matter. This enhances the gauge group of GR to include local Lorentz transformations. In view of a reduced phase space formulation of quantum gravity, in this work we completely gauge fix that Lorentz gauge symmetry by using a so-called triangular gauge. Having solved the Gauss constraints already classically opens access to new Hilbert space representations which are free of the complications that otherwise arise due to a non Abelian gauge symmetry. In that sense, a connection formulation as being pursued in Loop Quantum Gravity is no longer the only practicable option and other less dimension dependent representations e.g. based on triads and even metrics suggest themselves. These formulations make it easier to identify states representing non-degenerate quantum geometries and thus to investigate the hypersurface deformation algebra which implicitly assumes non-degeneracy.

Schr\"odinger operators often display singularities at the origin, the Coulomb problem in atomic physics or the various matter coupling terms in the Friedmann-Robertson-Walker problem being prominent examples. For various applications it would be desirable to have at one's disposal an explicit basis spanning a dense and invariant domain for such types of Schr\"odinger operators, for instance stationary perturbation theory or the Raleigh-Ritz method. Here we make the observation, that not only a such basis can indeed be provided but that in addition relevant matrix elements and inner products can be computed analytically in closed form, thus providing the required data e.g. for an analytical Gram-Schmid orthonormalisation., Comment: 9 pages

We propose algebraic protocols that enable the definition of protocol templates and session types analogous to the definition of domain-specific types with algebraic datatypes. Parameterized algebraic protocols subsume all regular as well as most context-free and nested session types and, at the same time, replace the expensive superlinear algorithms for type checking by a nominal check that runs in linear time. Algebraic protocols in combination with polymorphism increase expressiveness and modularity by facilitating new ways of parameterizing and composing session types.

This chapter focuses on the status of the implementation of the dynamics in the canonical version of Loop Quantum Gravity (LQG). Concretely this means to provide a mathematical meaning of the quantum Einstein equations, sometimes called Wheeler-DeWitt equations, to give a physical interpretation and Hilbert space structure to its solutions and to construct a representation of the algebra of observables including a physical Hamiltonian. This is a structural overview intentionally skipping technical details., Comment: 52 pages. This is a preprint of a chapter to appear in the "Handbook of Quantum Gravity", edited by Cosimo Bambi, Leonardo Modesto and Ilya Shapiro, 2023, Springer, reproduced with permission of Springer

All formalizations of session types rely on linear types for soundness as session-typed communication channels must change their type at every operation. Embedded language implementations of session types follow suit. They either rely on clever typing constructions to guarantee linearity statically, or on run-time checks that approximate linearity. We present a new language embedded implementation of session types, which is inspired by the inversion of control design principle. With our approach, all application programs are intrinsically session typed and unable to break linearity by construction. Linearity remains a proof obligation for a tiny encapsulated library that can be discharged once and for all when the library is built. We demonstrate that our proposed design extends to a wide range of features of session type systems: branching, recursion, multichannel and higher-order session, as well as context-free sessions. The multichannel extension provides an embedded implementation of session types which guarantees deadlock freedom by construction. The development reported in this paper is fully backed by type-checked Agda code., Comment: 27 pages; in submission

Session types provide a principled approach to typed communication protocols that guarantee type safety and protocol fidelity. Formalizations of session-typed communication are typically based on process calculi, concurrent lambda calculi, or linear logic. An alternative model based on context-sensitive typing and typestate has not received much attention due to its apparent restrictions. However, this model is attractive because it does not force programmers into particular patterns like continuation-passing style or channel-passing style, but rather enables them to treat communication channels like mutable variables. Polymorphic typestate is the key that enables a full treatment of session-typed communication. Previous work in this direction was hampered by its setting in a simply-typed lambda calculus. We show that higher-order polymorphism and existential types enable us to lift the restrictions imposed by the previous work, thus bringing the expressivity of the typestate-based approach on par with the competition. On this basis, we define PolyVGR, the system of polymorphic typestate for session types, establish its basic metatheory, type preservation and progress, and present a prototype implementation., Comment: 29 pages. Short version appears in PPDP 2023

Friction at water-carbon interfaces remains a major puzzle with theories and simulations unable to explain experimental trends in nanoscale waterflow. A recent theoretical framework -- quantum friction (QF)-- proposes to resolve these experimental observations by considering nonadiabatic coupling between dielectric fluctuations in water and graphitic surfaces. Here, using a classical model that enables fine-tuning of the solid's dielectric spectrum, we provide evidence from simulations in general support of QF. In particular, as features in the solid's dielectric spectrum begin to overlap with water's librational and Debye modes, we find an increase in friction in line with that proposed by QF. At the microscopic level, we find that this contribution to friction manifests more distinctly in the dynamics of the solid's charge density than that of water. Our findings suggest that experimental signatures of QF may be more pronounced in the solid's response rather than liquid water's.

The U(1)$^3$ model for 3+1 Euclidian signature general relativity is an interacting, generally covariant field theory with two physical polarisations that shares many features of Lorentzian general relativity. In particular, it displays a non-trivial realisation of the hypersurface deformation algebroid with non-trivial, i.e. phase space dependent structure functions rather than structure constants. In this paper we show that the model admits {\it an exact quantisation}. The quantisation rests on the observation that for this model and in the chosen representation of the canonical commutation relations the density unity hypersurface algebra {\it can be exponentiated on non-degenerate states}. These are states that represent a non-degenerate quantum metric and from a classical perspective are the relevant states on which the hypersurface algebra is representable. The representation of the algebra is exact, with no ambiguities involved and anomaly free. The quantum constraints can be exactly solved using {\it groupoid averaging} and the solutions admit a Hilbert space structure that agrees with the quantisation of a recently found reduced phase space formulation. Using the also recently found covariant action for that model, we start a path integral or spin foam formulation which, due to the Abelian character of the gauge group, is much simpler than for Lorentzian signature general relativity and provides an ideal testing ground for general spin foam models., Comment: 36p, no figures

A basic assumption in classical GR is that the metric field is nowhere degenerate in spacetime. In particular the induced metric on Cauchy surfaces must be nowhere degenerate. It is only under this assumption that one can derive the hypersurace deformation algebra between the initial value constraints which is absolutely transparent from the fact that the {\it inverse} of the induced metric is needed to close the algebra. This statement is independent of the density weight that one may want to equip the spatial metric with. Accordingly, the very definition of a non-anomalous representation of the hypersurface defomation algebra in quantum gravity has to address the issue of non-degenracy of the induced metric that is needed in the classical theory. In the Hilbert space representation employed in Loop Quantum Gravity (LQG) most emphasis has been layed to define an inverse metric operator on the dense domain of spin network states although they represent induced quantum geometries which are degenerate almost everywhere. It is no surprise that demonstration of closure of the constraint algebra on this domain meets difficulties because it is a sector of the quantum theory which is classically forbidden and which lies outside the domain of definition of the classical hypersurface deformation algebra. Various suggestions for addressing the issue such as non-standard operator topologies, dual spaces (habitats) and density weights have been propposed to address this issue with respect to the quantum dynamics of LQG. In this article we summarise these developments and argue that insisting on a dense domain of non-degenerate states within the LQG representation may provide a natural resolution of the issue thereby possibly avoiding the above mentioned non-standard constructions., Comment: 27p, no figures

In constructive quantum field theory (CQFT) it is customary to first regularise the theory at finite UV and IR cut-off. Then one first removes the UV cutoff using renormalisation techniques applied to families of CQFT's labelled by finite UV resolutions and then takes the thermodynamic limit. Alternatively, one may try to work directly without IR cut-off. More recently, wavelets have been proposed to define the renormalisation flow of CQFT's which is natural as they come accompanied with a multi-resolution analysis (MRA). However, wavelets so far have been mostly studied in the non-compact case. Practically useful wavelets that display compact support and some degree of smoothness can be constructed on the real line using Fourier space techniques but explicit formulae as functions of position are rarely available. Compactly supported wavelets can be periodised by summing over period translates keeping orthogonality properties but still yield to rather complicated expressions which generically lose their smoothness and position locality properties. It transpires that a direct approach to wavelets in the compact case is desirable. In this contribution we show that the Dirichlet-Shannon kernels serve as a natural scaling function to define generalised orthonormal wavelet bases on tori or copies of real lines respectively. These generalised wavelets are smooth, are simple explicitly computable functions, display quasi-local properties close to the Haar wavelet and have compact momentum supprt. Accordingly they have a built-in cut-off both in position and momentum, making them very useful for renormalisation applications., Comment: 20p, no figures

The Hamiltonian renormalisation programme motivated by constructive QFT and Osterwalder-Schrader reconstruction which was recently launched for bosonic field theories is extended to fermions. As fermion quantisation is not in terms of measures, the scheme has to be mildly modified accordingly. We exemplify the scheme for free fermions both for compact and non-compact spatial topologies respectively (i.e. with and without IR cut-off) and demonstrate that the convenient Dirichlet or Shannon coarse graining kernels recently advertised in a companion paper lead to a manifestly doubler free flow., Comment: 15p, no figures

Hamiltonian Renormalisation, as defined within this series of works, was derived from covariant Wilson renormalisation via Osterwalder-Schrader reconstruction. As such it directly applies to QFT with a true (physical) Hamiltonian bounded from below. The validity of the scheme was positively tested for free QFT in any dimension with or without Abelian gauge symmetries of Yang-Mills type. The aim of this Hamiltonian renormalisation scheme is to remove quantisation ambiguities of Hamiltonians in interacting QFT that remain even after UV and IR regulators are removed as it happens in highly non-linear QFT such as quantum gravity. While not derived for that case, the renormalisation flow formulae can without change also be applied to QFT without a single true Hamiltonian but rather an infinite number of Hamiltonian constraints. Then a central question is how the constraint algebra reacts to the renormalisation flow. This question should ultimately be addressed in quantum gravity. Before one considers this interacting, constrained QFT it is well motivated to consider a free, constrained QFT where the fixed point is explicitly known. In this paper we therefore address the case of parametrised field theory for which the quantum constraint algebra coincides simultaneously with the hypersurface deformation algebra of quantum gravity (or any other generally covariant theory) and the Virasoro algebra of free, closed, bosonic string theory or other CFT to which the results of this paper apply verbatim. The central result of our investigation is that finite resolution (discretised) constraint algebras {\it must not close} and that anomaly freeness of the continuum algebra is encoded in the convergence behaviour of the renormalisation flow., Comment: 39p, no figures

Java's type system mostly relies on type checking augmented with local type inference to improve programmer convenience. We study global type inference for Featherweight Generic Java (FGJ), a functional Java core language. Given generic class headers and field specifications, our inference algorithm infers all method types if classes do not make use of polymorphic recursion. The algorithm is constraint-based and improves on prior work in several respects. Despite the restricted setting, global type inference for FGJ is NP-complete., Comment: 33 pages, abridged version appears in ECOOP 2022

In order to determine the role of non-thermal transport of hot carriers which is decisive for the dissipation of energy in condensed matter we performed time-resolved broadband femtosecond transient reflectivity measurements on $7-197 \mathrm{nm}$ thick Bi(111) films epitaxially grown on Si(111). We monitored the behavior of the Fourier amplitude and the central frequency of the coherent $A_\mathrm{1g}$ phonon mode as function of the incident fluence, film thickness, and probe wavelength in the range of $580 -700 \mathrm{nm}$. The frequency redshift that follows photoexcitation was used as a robust quantity to determine the effective distribution of excited carriers that governs the displacive excitation mechanism of coherent $A_\mathrm{1g}$ phonons in Bi. For Bi films up to $50 \mathrm{nm}$ thickness a homogeneous excitation due to the ultrafast transport of hot charge carriers is observed, limited by a carrier penetration depth of $60 \mathrm{nm}$ independent of the totally deposited laser energy., Comment: 7 pages, 2 figures

A two-stage lightweight online dereverberation algorithm for hearing devices is presented in this paper. The approach combines a multi-channel multi-frame linear filter with a single-channel single-frame post-filter. Both components rely on power spectral density (PSD) estimates provided by deep neural networks (DNNs). By deriving new metrics analyzing the dereverberation performance in various time ranges, we confirm that directly optimizing for a criterion at the output of the multi-channel linear filtering stage results in a more efficient dereverberation as compared to placing the criterion at the output of the DNN to optimize the PSD estimation. More concretely, we show that training this stage end-to-end helps further remove the reverberation in the range accessible to the filter, thus increasing the \textit{early-to-moderate} reverberation ratio. We argue and demonstrate that it can then be well combined with a post-filtering stage to efficiently suppress the residual late reverberation, thereby increasing the \textit{early-to-final} reverberation ratio. This proposed two stage procedure is shown to be both very effective in terms of dereverberation performance and computational demands, as compared to e.g. recent state-of-the-art DNN approaches. Furthermore, the proposed two-stage system can be adapted to the needs of different types of hearing-device users by controlling the amount of reduction of early reflections., Comment: Accepted for publication in EURASIP Journal on Audio, Speech and Music Processing