Publications

Deep neural networks can learn powerful prior probability models for images, as evidenced by the high-quality generations obtained with recent score-based diffusion methods. But the means by which these networks capture complex global statistical structure, apparently without suffering from the curse of dimensionality, remain a mystery. To study this, we incorporate diffusion methods into a multi-scale decomposition, reducing dimensionality by assuming a stationary local Markov model for wavelet coefficients conditioned on coarser-scale coefficients. We instantiate this model using convolutional neural networks (CNNs) with local receptive fields, which enforce both the stationarity and Markov properties. Global structures are captured using a CNN with receptive fields covering the entire (but small) low-pass image. We test this model on a dataset of face images, which are highly non-stationary and contain large-scale geometric structures.
Remarkably, denoising, super-resolution, and image synthesis results all demonstrate that these structures can be captured with significantly smaller conditioning neighborhoods than required by a Markov model implemented in the pixel domain. Our results show that score estimation for large complex images can be reduced to low-dimensional Markov conditional models across scales, alleviating the curse of dimensionality.

The accurate computation of forces and other energy derivatives has been a long-standing challenge for quantum Monte Carlo methods. A number of technical obstacles contribute to this challenge. We discuss how these obstacles can be removed with the auxiliary-field quantum Monte Carlo (AFQMC) approach. AFQMC is a general, high-accuracy, many-body total-energy method for molecules and solids. The implementation of back-propagation for pure estimators allows direct calculation of gradients of the energy via the Hellmann-Feynman theorem. A planewave basis with norm-conserving pseudopotentials is used for the study of periodic bulk materials. Completeness of the planewave basis minimizes the effect of so-called Pulay terms. The ionic pseudopotentials, which can be incorporated in AFQMC in exactly the same manner as in standard independent-electron methods, regulate the force and stress estimators and eliminate any potential divergence of the Monte Carlo variances. The resulting approach allows applications of full geometry optimizations in bulk materials. It also paves the way for many-body computations of the phonon spectrum in solids.

Using a density-functional theory plus dynamical mean-field theory methodology, we compute the many-body electronic structure and optical conductivity of NdNiO

Understanding carrier trapping in solids has proven key to semiconductor technologies but observations thus far have relied on ensembles of point defects, where the impact of neighboring traps or carrier screening is often important. Here, we investigate the capture of photo-generated holes by an individual negatively-charged nitrogen-vacancy (NV) center in diamond at room temperature. Using an externally gated potential to minimize space-charge effects, we find the capture probability under electric fields of variable sign and amplitude shows an asymmetric-bell-shaped response with maximum at zero voltage. To interpret these observations, we run semi-classical Monte Carlo simulations modeling carrier trapping through a cascade process of phonon emission, and obtain electric-field-dependent capture probabilities in good agreement with experiment. Since the mechanisms at play are insensitive to the trap characteristics, the capture cross sections we observe - largely exceeding those derived from ensemble measurements - should also be present in materials platforms other than diamond.

Electromagnetic fields can be confined in the presence of metal nanostructures. Recently, subnanometer scale confinement has been demonstrated to occur at atomic protrusions on plasmonic nanostructures. Such an extreme field may dominate atomic-scale light--matter interactions in

The microscopic mechanism of the light-matter interactions that induce orbital angular momentum (OAM) in electromagnetic fields is not thoroughly understood. In this work, we employ Archimedean spiral vortex generators in time-resolved numerical simulations using the Octopus code to observe the behind-the-scenes of OAM generation. We send a perfect circularly-polarized plane-wave light onto plasmonic optical vortex generators and observe the resulting twisted light formation with complete spatio-temporal information. In agreement with previous works, we find that emission from the plasmonic spiral branches shapes the vortex-like structure and governs the OAM generation in the outgoing electromagnetic field. To characterize the generated beam further, we emulate the emission from vortex generators with current emitters preserving the spiral geometry. We subject a point-particle system to the generated field and record the orbital angular momentum transfer between the electromagnetic field and the point particle. Finally, we probe the OAM density locally by studying the induced classical trajectory of point particles, which provides further insight into the spatio-temporal features of the induced OAM.

The electric field of an intense laser pulse can directly modify the electronic properties of materials via electromodulation up to the petahertz regime. In this regime, the energy of the quiver motion of the electron-hole pair is comparable with the photon energy, which results in complex nonadiabatic dynamics. This regime opens opportunities to probe the electronic structure of materials on the attosecond timescale. Here, we show how the quasistatic electromodulation spectroscopy based on the Franz-Keldysh effect (FKE) connects with its nonadiabatic limit, which we find to be determined by resonant transitions between Floquet-Bloch states. This insight can be applied to measure the effective mass, ponderomotive and binding energies of the electron-hole pair on a few-femtosecond timescale. We demonstrate this by experimentally investigating laser-field-driven fused silica, a prototypical material for light-wave electronics, with extreme-ultraviolet attosecond pulse trains. We reproduce the experimental transient absorption spectra with an effective band structure and a dynamical Franz-Keldysh model, offering a simple parametrization for a theoretically challenging but technologically abundant material. Ab initio calculations in α-quartz highlight the contributions of specific bands, symmetry, and crystal orientation that are hidden in the experimental data due to randomized crystallographic orientation and finite temporal and spatial coherence. We show that the dynamical FKE can be explained as a third-order nonlinear process in the weak-field regime. The delay-dependent position of the absorption maxima and minima has a minimum tilt angle, determined by transitions between the underlying Floquet-Bloch states. In our analysis, we discuss the main experimental observables and show their connection to the parameters of the solid, providing the basis for nonadiabatic electromodulation spectroscopy.

For the past half-century, structural biologists relied on the notion that similar protein sequences give rise to similar structures and functions. While this assumption has driven research to explore certain parts of the protein universe, it disregards spaces that don’t rely on this assumption. Here we explore areas of the protein universe where similar protein functions can be achieved by different sequences and different structures. We predict ~200,000 structures for diverse protein sequences from 1,003 representative genomes across the microbial tree of life and annotate them functionally on a per-residue basis. Structure prediction is accomplished using the World Community Grid, a large-scale citizen science initiative. The resulting database of structural models is complementary to the AlphaFold database, with regards to domains of life as well as sequence diversity and sequence length. We identify 148 novel folds and describe examples where we map specific functions to structural motifs. We also show that the structural space is continuous and largely saturated, highlighting the need for a shift in focus across all branches of biology, from obtaining structures to putting them into context and from sequence-based to sequence-structure-function based meta-omics analyses.

Randomized sketches of a tensor product of pvectors follow a tradeoff between statistical efficiency and computational acceleration. Commonly used approaches avoid computing the high-dimensional tensor product explicitly, resulting in a suboptimal dependence of O(3p) in the embedding dimension. We propose a simple Complex-to-Real (CtR) modification of well-known sketches that replaces real random projections by complex ones, incurring a lower O(2p)factor in the embedding dimension. The output of our sketches is real-valued, which renders
their downstream use straightforward. In particular, we apply our sketches to p-fold self-tensored inputs corresponding to the feature maps of the
polynomial kernel. We show that our method achieves state-of-the-art performance in terms of accuracy and speed compared to other randomized approximations from the literature.

This paper reviews the growing field of Bayesian prediction. Bayes point and interval prediction are defined and exemplified and situated in statistical prediction more generally. Then, four general approaches to Bayes prediction are defined and we turn to predictor selection. This can be done predictively or non-predictively and predictors can be based on single models or multiple models. We call these latter cases unitary predictors and model average predictors, respectively. Then we turn to the most recent aspect of prediction to emerge, namely prediction in the context of large observational data sets and discuss three further classes of techniques. We conclude with a summary and statement of several current open problems.