Search

In recent years, action-detected ultrafast spectroscopies have gained popularity. These approaches offer some advantages over their coherently-detected counterparts, enabling spatially-resolved and operando measurements with high sensitivity. However, there are also fundamental limitations connected to the different process of signal generation in action-detected experiments. Specifically, state mixing by nonlinear interactions during signal emission leads to a large static background which can obscure the excited-state dynamics. This could severely limit the applicability of action-detected spectroscopy to study energy transfer in larger systems. Here we perform fluorescence-detected two-dimensional electronic spectroscopy (F-2DES) of the light-harvesting II (LH2) complex from purple bacteria. We demonstrate that the B800-B850 energy transfer process in LH2 is barely discernable in F-2DES, representing a ~6.2% rise of the lower cross-peak intensity. This is in stark contrast to measurements using coherently-detected 2DES where the lower cross-peak reveals energy transfer with 100% contrast. We explain the weak presence of excited-state dynamics using a disordered excitonic model with realistic experimental conditions. We further derive a general formula for the presence of excited-state signals in multi-chromophoric aggregates, dependent on the aggregate geometry and size, and the interplay of excitonic coupling and disorder. We find that, dependent on the excitonic state structure, the excited state dynamics in F-2DES can be visible even in large aggregates. Our work shows that the signatures of energy transfer in F-2DES can be used to directly infer the excitonic structure in multichromophoric systems.

Apportionment is the task of assigning resources to entities with different entitlements in a fair manner, and specifically a manner that is as proportional as possible. The best-known application concerns the assignment of parliamentary seats to political parties based on their share in the popular vote. Here we enrich the standard model of apportionment by associating each seat with a weight that reflects the value of that seat, for example because seats come with different roles, such as chair or treasurer, that have different (objective) values. We define several apportionment methods and natural fairness requirements for this new setting, and study the extent to which our methods satisfy our requirements. Our findings show that full fairness is harder to achieve than in the standard apportionment setting. At the same time, for relaxations of those requirements we can achieve stronger results than in the more general model of weighted fair division, where the values of objects are subjective.

We develop a new functional-analytic technique for investigating the degree of noncompactness of an operator defined on a quasinormed space and taking values in a Marcinkiewicz space. The main result is a general principle from which it can be derived that such operators are almost always maximally noncompact in the sense that their ball measure of noncompactness coincides with their operator norm. We point out specifications of the universal principle to the case of the identity operator., Comment: 22 pages

Rotary atomization is used in a wide variety of fields, exploiting the external control option of the spray while no high-pressure fluid is needed. Most papers on rotary atomization deal with liquid jet breakup, while external spray characteristics are rarely evaluated; this is performed currently. The water spray was measured by a two-component Phase Doppler Anemometer. The optical setup requires a special measurement chamber to avoid spray deposition on the optical components. Therefore, the first goal was to find a proper filter that enables the removal of biased droplets by secondary flows. Since most droplets have a similar radial-to-tangential velocity ratio at each measurement point, i.e., scattering around a line, this was the first component of the best filter. The second component was the need for a positive radial velocity component. This filter efficiently removed droplets originating from alternative processes, increasing the R2 of the line fit. The physical soundness of this filter was checked by evaluating the effect of filtering on the angle of the velocity components of each droplet at a given measurement point. The proposed filter efficiently detected recirculation, a secondary effect of the measurement setup with less regular data set shapes. Finally, the slope and intercept values of the fitted lines were evaluated and presented. The mean of the former followed the same trend irrespective of the rotational speed and the mass flow rate; it was principally dependent on the radial distance from the atomizer. The intercept showed a regular but less universal behavior.

Online discussion platforms are a vital part of the public discourse in a deliberative democracy. However, how to interpret the outcomes of the discussions on these platforms is often unclear. In this paper, we propose a novel and explainable method for selecting a set of most representative, consistent points of view by combining methods from computational social choice and abstract argumentation. Specifically, we model online discussions as abstract argumentation frameworks combined with information regarding which arguments voters approve of. Based on ideas from approval-based multiwinner voting, we introduce several voting rules for selecting a set of preferred extensions that represents voters' points of view. We compare the proposed methods across several dimensions, theoretically and in numerical simulations, and give clear suggestions on which methods to use depending on the specific situation., Comment: 33 pages. Extended version of an accepted AAMAS-24 paper

The core is a strong fairness notion in multiwinner voting and participatory budgeting (PB). It is known that the core can be empty if we consider cardinal utilities, but it is not known whether it is always satisfiable with approval-ballots. In this short note, I show that in approval-based PB the core can be empty for nearly all satisfaction functions that are based on the cost of a project. In particular, I show that the core can be empty for the cost satisfaction function, satisfaction functions based on diminishing marginal returns and the share. However, it remains open whether the core can be empty for the cardinality satisfaction function.

Fluorescence lifetime imaging microscopy (FLIM) is a well-established technique with numerous imaging applications. Yet, one of the limitations of FLIM is that it provides information about the emitting state only. Here, we present an extension of FLIM by interferometric measurement of fluorescence excitation spectra. Interferometric Excitation Fluorescence Lifetime Imaging Microscopy (ixFLIM) reports on the correlation of the excitation spectra and emission lifetime, providing the correlation between the ground-state absorption and excited-state emission. As such, it extends the applicability of FLIM and removes some of its limitations. We introduce ixFLIM on progressively more complex systems and apply it to quantitative resonance energy transfer imaging from a single measurement., Comment: 29 pages (19 main and 10 supporting), 18 figures

Voting in multi-issue domains allows for compromise outcomes that satisfy all voters to some extent. Such fairness considerations, however, open the possibility of a special form of manipulation: free-riding. By untruthfully opposing a popular opinion in one issue, voters can receive increased consideration in other issues. We study under which conditions this is possible. Additionally, we study free-riding from a computational and experimental point of view. Our results show that free-riding in multi-issue domains is largely unavoidable, but comes at a non-negligible individual risk for voters. Thus, the allure of free-riding is smaller than one could intuitively assume., Comment: This is an extended version of the publication with the same name in the proceedings of AAMAS 2023

Controlling the electron dynamics in matter by individual oscillations of light fields has recently led to the development of attosecond metrology. One of the phenomena resulting from the nonlinear response of materials in the strong-field interaction regime is coherent emission of high energy photons. High harmonic spectra carry the fingerprints of sub-cycle electronic motion and the energy structure of the studied system. Here we show that tailoring the waveform of the driving light by using a coherent combination with its third harmonic frequency allows to control the electron tunneling time within each half-cycle of the fundamental wave with attosecond precision. We introduce an experimental scheme in which we simultaneously monitor the modulation of amplitude and emission delays of high harmonic radiation and the excited electron population generated in crystalline silicon as a function of the relative phase between the {\omega}-3{\omega} fields. The results reveal unambiguously the connection between the dynamics of electron tunneling and high harmonic generation processes in solids.

In this work, we analyze the relation between reparametrizations of gradient flow and the induced implicit bias in linear models, which encompass various basic regression tasks. In particular, we aim at understanding the influence of the model parameters - reparametrization, loss, and link function - on the convergence behavior of gradient flow. Our results provide conditions under which the implicit bias can be well-described and convergence of the flow is guaranteed. We furthermore show how to use these insights for designing reparametrization functions that lead to specific implicit biases which are closely connected to $\ell_p$- or trigonometric regularizers.

We consider covariance estimation of any subgaussian distribution from finitely many i.i.d. samples that are quantized to one bit of information per entry. Recent work has shown that a reliable estimator can be constructed if uniformly distributed dithers on $[-\lambda,\lambda]$ are used in the one-bit quantizer. This estimator enjoys near-minimax optimal, non-asymptotic error estimates in the operator and Frobenius norms if $\lambda$ is chosen proportional to the largest variance of the distribution. However, this quantity is not known a-priori, and in practice $\lambda$ needs to be carefully tuned to achieve good performance. In this work we resolve this problem by introducing a tuning-free variant of this estimator, which replaces $\lambda$ by a data-driven quantity. We prove that this estimator satisfies the same non-asymptotic error estimates - up to small (logarithmic) losses and a slightly worse probability estimate. We also show that by using refined data-driven dithers that vary per entry of each sample, one can construct an estimator satisfying the same estimation error bound as the sample covariance of the samples before quantization -- again up logarithmic losses. Our proofs rely on a new version of the Burkholder-Rosenthal inequalities for matrix martingales, which is expected to be of independent interest.

We propose a new algorithm for the problem of recovering data that adheres to multiple, heterogeneous low-dimensional structures from linear observations. Focusing on data matrices that are simultaneously row-sparse and low-rank, we propose and analyze an iteratively reweighted least squares (IRLS) algorithm that is able to leverage both structures. In particular, it optimizes a combination of non-convex surrogates for row-sparsity and rank, a balancing of which is built into the algorithm. We prove locally quadratic convergence of the iterates to a simultaneously structured data matrix in a regime of minimal sample complexity (up to constants and a logarithmic factor), which is known to be impossible for a combination of convex surrogates. In experiments, we show that the IRLS method exhibits favorable empirical convergence, identifying simultaneously row-sparse and low-rank matrices from fewer measurements than state-of-the-art methods. Code is available at https://github.com/ckuemmerle/simirls., Comment: 35 pages, 7 figures

Flare frequency distributions represent a key approach to addressing one of the largest problems in solar and stellar physics: determining the mechanism that counter-intuitively heats coronae to temperatures that are orders of magnitude hotter than the corresponding photospheres. It is widely accepted that the magnetic field is responsible for the heating, but there are two competing mechanisms that could explain it: nanoflares or Alfv\'en waves. To date, neither can be directly observed. Nanoflares are, by definition, extremely small, but their aggregate energy release could represent a substantial heating mechanism, presuming they are sufficiently abundant. One way to test this presumption is via the flare frequency distribution, which describes how often flares of various energies occur. If the slope of the power law fitting the flare frequency distribution is above a critical threshold, $\alpha=2$ as established in prior literature, then there should be a sufficient abundance of nanoflares to explain coronal heating. We performed $>$600 case studies of solar flares, made possible by an unprecedented number of data analysts via three semesters of an undergraduate physics laboratory course. This allowed us to include two crucial, but nontrivial, analysis methods: pre-flare baseline subtraction and computation of the flare energy, which requires determining flare start and stop times. We aggregated the results of these analyses into a statistical study to determine that $\alpha = 1.63 \pm 0.03$. This is below the critical threshold, suggesting that Alfv\'en waves are an important driver of coronal heating., Comment: 1,002 authors, 14 pages, 4 figures, 3 tables, published by The Astrophysical Journal on 2023-05-09, volume 948, page 71

In this survey, we review the literature investigating participatory budgeting as a social choice problem. Participatory Budgeting (PB) is a democratic process in which citizens are asked to vote on how to allocate a given amount of public money to a set of projects. From a social choice perspective, it corresponds then to the problem of aggregating opinions about which projects should be funded, into a budget allocation satisfying a budget constraint. This problem has received substantial attention in recent years and the literature is growing at a fast pace. In this survey, we present the most important research directions from the literature, each time presenting a large set of representative results. We only focus on the indivisible case, that is, PB problems in which projects can either be fully funded or not at all. The aim of the survey is to present a comprehensive overview of the state of the research on PB. We aim at providing both a general overview of the main research questions that are being investigated, and formal and unified definitions of the most important technical concepts from the literature.

The ability to measure the satisfaction of (groups of) voters is a crucial prerequisite for formulating proportionality axioms in approval-based participatory budgeting elections. Two common - but very different - ways to measure the satisfaction of a voter consider (i) the number of approved projects and (ii) the total cost of approved projects, respectively. In general, it is difficult to decide which measure of satisfaction best reflects the voters' true utilities. In this paper, we study proportionality axioms with respect to large classes of approval-based satisfaction functions. We establish logical implications among our axioms and related notions from the literature, and we ask whether outcomes can be achieved that are proportional with respect to more than one satisfaction function. We show that this is impossible for the two commonly used satisfaction functions when considering proportionality notions based on extended justified representation, but achievable for a notion based on proportional justified representation. For the latter result, we introduce a strengthening of priceability and show that it is satisfied by several polynomial-time computable rules, including the Method of Equal Shares and Phragm\`en's sequential rule.

As the array dimension of massive MIMO systems increases to unprecedented levels, two problems occur. First, the spatial stationarity assumption along the antenna elements is no longer valid. Second, the large array size results in an unacceptably high power consumption if high-resolution analog-to-digital converters are used. To address these two challenges, we consider a Bussgang linear minimum mean square error (BLMMSE)-based channel estimator for large scale massive MIMO systems with one-bit quantizers and a spatially non-stationary channel. Whereas other works usually assume that the channel covariance is known at the base station, we consider a plug-in BLMMSE estimator that uses an estimate of the channel covariance and rigorously analyze the distortion produced by using an estimated, rather than the true, covariance. To cope with the spatial non-stationarity, we introduce dithering into the quantized signals and provide a theoretical error analysis. In addition, we propose an angular domain fitting procedure which is based on solving an instance of non-negative least squares. For the multi-user data transmission phase, we further propose a BLMMSE-based receiver to handle one-bit quantized data signals. Our numerical results show that the performance of the proposed BLMMSE channel estimator is very close to the oracle-aided scheme with ideal knowledge of the channel covariance matrix. The BLMMSE receiver outperforms the conventional maximum-ratio-combining and zero-forcing receivers in terms of the resulting ergodic sum rate., Comment: submitted to IEEE Transactions on Communications

In this short note, we propose a new method for quantizing the weights of a fully trained neural network. A simple deterministic pre-processing step allows us to quantize network layers via memoryless scalar quantization while preserving the network performance on given training data. On one hand, the computational complexity of this pre-processing slightly exceeds that of state-of-the-art algorithms in the literature. On the other hand, our approach does not require any hyper-parameter tuning and, in contrast to previous methods, allows a plain analysis. We provide rigorous theoretical guarantees in the case of quantizing single network layers and show that the relative error decays with the number of parameters in the network if the training data behaves well, e.g., if it is sampled from suitable random distributions. The developed method also readily allows the quantization of deep networks by consecutive application to single layers.

We present an elementary proof of a well-known theorem of Cheeger which states that if a metric-measure space $X$ supports a $p$-Poincar\'e inequality, then the $N^{1,p}(X)$ Sobolev space is reflexive and separable whenever $p\in (1,\infty)$. We also prove separability of the space when $p=1$. Our proof is based on a straightforward construction of an equivalent norm on $N^{1,p}(X)$, $p\in [1,\infty)$, that is uniformly convex when $p\in (1,\infty)$. Finally, we explicitly construct a functional that is pointwise comparable to the minimal $p$-weak upper gradient, when $p\in (1,\infty)$.

One of the main complications for the interpretation of reflectance spectra of airless planetary bodies is surface alteration by space weathering caused by irradiation by solar wind and micrometeoroid particles. We aim to evaluate the damage to the samples from H and laser irradiation and relate it to the observed alteration in the spectra. We used olivine (OL) and pyroxene (OPX) pellets irradiated by 5 keV H ions and individual fs laser pulses and measured their visible (VIS) and near-infrared (NIR) spectra. We observed the pellets with scanning and transmission electron microscopy. We studied structural, mineralogical, and chemical modifications in the samples and connected them to changes in the reflectance spectra. In both minerals, H irradiation induces partially amorphous sub-surface layers containing small vesicles. In OL pellets, these vesicles are more tightly packed than in OPX ones. Related spectral change is mainly in the VIS spectral slope. Changes due to laser irradiation are mostly dependent on material's melting temperature. Only the laser-irradiated OL contains nanophase Fe particles, which induce detectable spectral slope change throughout the measured spectral range. Our results suggest that spectral changes at VIS-NIR wavelengths are mainly dependent on thickness of (partially) amorphous sub-surface layers. Amorphisation smooths microroughness, increasing the contribution of volume scattering and absorption over surface scattering. Soon after exposure to the space environment, the appearance of partially amorphous sub-surface layers results in rapid changes in the VIS spectral slope. In later stages (onset of micrometeoroid bombardment), we expect an emergence of nanoparticles to also mildly affect the NIR spectral slope. An increase in dimensions of amorphous layers and vesicles in the more space-weathered material will only cause band-depth variation and darkening., Comment: 9 pages, 8 figures, 1 table. Abstract was modified to comply with the required maximum number of characters of arXiv abstracts

In many applications, solutions of numerical problems are required to be non-negative, e.g., when retrieving pixel intensity values or physical densities of a substance. In this context, non-negative least squares (NNLS) is a ubiquitous tool, e.g., when seeking sparse solutions of high-dimensional statistical problems. Despite vast efforts since the seminal work of Lawson and Hanson in the '70s, the non-negativity assumption is still an obstacle for the theoretical analysis and scalability of many off-the-shelf solvers. In the different context of deep neural networks, we recently started to see that the training of overparametrized models via gradient descent leads to surprising generalization properties and the retrieval of regularized solutions. In this paper, we prove that, by using an overparametrized formulation, NNLS solutions can reliably be approximated via vanilla gradient flow. We furthermore establish stability of the method against negative perturbations of the ground-truth. Our simulations confirm that this allows the use of vanilla gradient descent as a novel and scalable numerical solver for NNLS. From a conceptual point of view, our work proposes a novel approach to trading side-constraints in optimization problems against complexity of the optimization landscape, which does not build upon the concept of Lagrangian multipliers.

This paper presents the Standalone Neural ODE (sNODE), a continuous-depth neural ODE model capable of describing a full deep neural network. This uses a novel nonlinear conjugate gradient (NCG) descent optimization scheme for training, where the Sobolev gradient can be incorporated to improve smoothness of model weights. We also present a general formulation of the neural sensitivity problem and show how it is used in the NCG training. The sensitivity analysis provides a reliable measure of uncertainty propagation throughout a network, and can be used to study model robustness and to generate adversarial attacks. Our evaluations demonstrate that our novel formulations lead to increased robustness and performance as compared to ResNet models, and that it opens up for new opportunities for designing and developing machine learning with improved explainability., Comment: 25 pages, 15 figures; typos corrected

We introduce a family of normative principles to assess fairness in the context of participatory budgeting. These principles are based on the fundamental idea that budget allocations should be fair in terms of the resources invested into meeting the wishes of individual voters. This is in contrast to earlier proposals that are based on specific assumptions regarding the satisfaction of voters with a given budget allocation. We analyse these new principles in axiomatic, algorithmic, and experimental terms.

Increasing the speed limits of conventional electronics requires innovative approaches to manipulate other quantum properties of electrons besides their charge. An alternative approach utilizes the valley degree of freedom in low-dimensional semiconductors. Here we demonstrate that the valley degeneracy of exciton energies in transition metal dichalcogenide monolayers may be lifted by coherent optical interactions on timescales corresponding to few tens of femtoseconds. The optical Stark and Bloch-Siegert effects generated by strong nonresonant circularly-polarized light induce valley-selective blue shifts of exciton quantum levels by more than 30 meV. We show these phenomena by studying the two most intensive exciton resonances in transiton metal dichalcogenide monolayers and compare the results to a theoretical model, which properly includes the Coulomb interaction and exciton dispersion. These results open the door for ultrafast valleytronics working at multiterahertz frequencies., Comment: 6 pages, 3 figures, published in npj 2D Materials and Applications

We report on the theoretical and experimental investigation of valley-selective optical Stark and Bloch-Siegert shifts of exciton resonances in monolayers WSe$_2$ and MoS$_2$ induced by strong circularly polarized nonresonant optical fields. We predict and observe transient shifts of both 1sA and 1sB exciton transitions in the linear interaction regime. The theoretical description is based on semiconductor Bloch equations. The solutions of the equations are obtained with a modified perturbation technique, which takes into account many-body Coulomb interaction effects. These solutions allow to explain the polarization dependence of the shifts and calculate their values analytically. We found experimentally the limits of the applicability of the theoretical description by observing the transient exciton spectra change due to many-body effects at high field amplitudes of the driving wave., Comment: 24 pages, 10 figures

The inverse problem of supervised reconstruction of depth-variable (time-dependent) parameters in a neural ordinary differential equation (NODE) is considered, that means finding the weights of a residual network with time continuous layers. The NODE is treated as an isolated entity describing the full network as opposed to earlier research, which embedded it between pre- and post-appended layers trained by conventional methods. The proposed parameter reconstruction is done for a general first order differential equation by minimizing a cost functional covering a variety of loss functions and penalty terms. A nonlinear conjugate gradient method (NCG) is derived for the minimization. Mathematical properties are stated for the differential equation and the cost functional. The adjoint problem needed is derived together with a sensitivity problem. The sensitivity problem can estimate changes in the network output under perturbation of the trained parameters. To preserve smoothness during the iterations the Sobolev gradient is calculated and incorporated. As a proof-of-concept, numerical results are included for a NODE and two synthetic datasets, and compared with standard gradient approaches (not based on NODEs). The results show that the proposed method works well for deep learning with infinite numbers of layers, and has built-in stability and smoothness., Comment: 26 pages, 3 figures

In deep learning it is common to overparameterize neural networks, that is, to use more parameters than training samples. Quite surprisingly training the neural network via (stochastic) gradient descent leads to models that generalize very well, while classical statistics would suggest overfitting. In order to gain understanding of this implicit bias phenomenon we study the special case of sparse recovery (compressed sensing) which is of interest on its own. More precisely, in order to reconstruct a vector from underdetermined linear measurements, we introduce a corresponding overparameterized square loss functional, where the vector to be reconstructed is deeply factorized into several vectors. We show that, if there exists an exact solution, vanilla gradient flow for the overparameterized loss functional converges to a good approximation of the solution of minimal $\ell_1$-norm. The latter is well-known to promote sparse solutions. As a by-product, our results significantly improve the sample complexity for compressed sensing via gradient flow/descent on overparameterized models derived in previous works. The theory accurately predicts the recovery rate in numerical experiments. Our proof relies on analyzing a certain Bregman divergence of the flow. This bypasses the obstacles caused by non-convexity and should be of independent interest.

Let $\Omega,\Omega'\subset\mathbb{R}^3$ be Lipschitz domains, let $f_m:\Omega\to\Omega'$ be a sequence of homeomorphisms with prescribed Dirichlet boundary condition and $\sup_m \int_{\Omega}(|Df_m|^2+1/J^2_{f_m})<\infty$. Let $f$ be a weak limit of $f_m$ in $W^{1,2}$. We show that $f$ is invertible a.e., more precisely it satisfies the (INV) condition of Conti and De Lellis and thus it has all the nice properties of mappings in this class. Generalization to higher dimensions and an example showing sharpness of the condition $1/J^2_f\in L^1$ are also given. Using this example we also show that unlike the planar case the class of weak limits and the class of strong limits of $W^{1,2}$ Sobolev homeomorphisms in $\mathbb{R}^3$ are not the same.

Distributed combustion, often associated with the low-oxygen condition, offers ultra-low NOX emission. However, it was recently achieved without combustion air dilution or internal flue gas recirculation, using a distinct approach called Mixture Temperature-Controlled combustion. Here, the fuel-air stream is cooled at the inlet to delay ignition and hence foster homogeneous mixture formation. The aim of this numerical study aims to understand the operation of this combustion concept better and present a robust framework for distributed combustion modeling in a parameter range where such operation was not predicted before by any existing theory. Further, liquid fuel combustion was evaluated that brings additional complexity. Four operating conditions were presented at which distributed combustion was observed. The reacting flow was modeled by Flamelet-Generated Manifold, based on a detailed n-dodecane mechanism. The Zimont turbulent flame speed model was used with significantly reduced coefficients to achieve distributed combustion. The droplets of airblast atomization were tracked in a Lagrangian frame. The numerical results were validated by Schlieren images and acoustic spectra. It was concluded that the reactant dilution ratio remained below 0.25 through the combustion chamber, revealing that the homogeneous fuel-air mixture is the principal reason for excellent flame stability and ultra-low NOX emission without significant internal recirculation. The potential applications of these results are boilers, furnaces, and gas turbines.