Search for: All records

Score matching is an approach to learning probability distributions parametrized up to a constant of proportionality (e.g. Energy-Based Models). The idea is to fit the score of the distribution, rather than the likelihood, thus avoiding the need to evaluate the constant of proportionality. While there's a clear algorithmic benefit, the statistical "cost'' can be steep: recent work by Koehler et al. 2022 showed that for distributions that have poor isoperimetric properties (a large Poincaré or log-Sobolev constant), score matching is substantially statistically less efficient than maximum likelihood. However, many natural realistic distributions, e.g. multimodal distributions as simple as a mixture of two Gaussians in one dimension -- have a poor Poincaré constant. 

Score matching is an approach to learning probability distributions parametrized up to a constant of proportionality (e.g. Energy-Based Models). The idea is to fit the score of the distribution, rather than the likelihood, thus avoiding the need to evaluate the constant of proportionality. While there's a clear algorithmic benefit, the statistical "cost'' can be steep: recent work by Koehler et al. 2022 showed that for distributions that have poor isoperimetric properties (a large Poincaré or log-Sobolev constant), score matching is substantially statistically less efficient than maximum likelihood. However, many natural realistic distributions, e.g. multimodal distributions as simple as a mixture of two Gaussians in one dimension -- have a poor Poincaré constant. In this paper, we show a close connection between the mixing time of a broad class of Markov processes with generator  and an appropriately chosen generalized score matching loss that tries to fit pp. This allows us to adapt techniques to speed up Markov chains to construct better score-matching losses. In particular, ``preconditioning'' the diffusion can be translated to an appropriate ``preconditioning'' of the score loss. Lifting the chain by adding a temperature like in simulated tempering can be shown to result in a Gaussian-convolution annealed score matching loss, similar to Song and Ermon, 2019. Moreover, we show that if the distribution being learned is a finite mixture of Gaussians in d dimensions with a shared covariance, the sample complexity of annealed score matching is polynomial in the ambient dimension, the diameter of the means, and the smallest and largest eigenvalues of the covariance -- obviating the Poincaré constant-based lower bounds of the basic score matching loss shown in Koehler et al. 2022. 

We explore the properties of interferometric data from high-redshift 21 cm measurements using the Murchison Widefield Array (MWA). These data contain the redshifted 21 cm signal, contamination from continuum foreground sources, and radiometric noise. The 21 cm signal from the Epoch of Reionization (EoR) is expected to be highly Gaussian, which motivates the use of the power spectrum as an effective statistical tool for extracting astrophysical information. We find that foreground contamination introduces non-Gaussianity into the distribution of measurements and then use this information to separate Gaussian from the non-Gaussian signal. We present improved upper limits on the 21 cm EoR power spectrum from the MWA using a Gaussian component of the data, based on the existing analysis from C. D. Nunhokee et al. 2025. This is extracted as the best-fitting Gaussian to the measured data. Our best 2σ (thermal+sample variance) limit for 268 hr of data improves from (30.2 mK)2 to (23.0 mK)2 at z = 6.5 for the East–West polarization, and from (39.2 mK)2 to (21.7 mK)2 = 470 mK2 in North–South. The best limits at z = 6.8 (z = 7.0) improve to P < (25.9 mK)2 (P < (32.0 mK)2) and k = 0.18h Mpc‑1 (k = 0.21h Mpc‑1). Results are compared with realistic simulations, which indicate that leakage from foreground contamination is a source of the non-Gaussian behavior. 

This paper presents the spherically averaged 21 cm power spectrum derived from Epoch of Reionization (EoR) observations conducted with the Murchison Widefield Array (MWA). The analysis uses EoR0-field data, centered at (R.A. = 0h, decl. = ‑27∘), collected between 2013 and 2023. Building on the improved methodology described in C. M. Trott et al. (2024), we incorporate additional data quality control techniques introduced in C. D. Nunhokee (2020). We report the lowest-power-level limits on the EoR power spectrum at redshifts z = 6.5, z = 6.8, and z = 7.0. These power levels, measured in the east–west polarization, are (30.2)2 mK2 at k = 0.18 h Mpc‑1, (31.2)2 mK2 at k = 0.18 h Mpc‑1, and (39.1)2 mK2 at k = 0.21 h Mpc‑1, respectively. The total integration time amounts to 268 hr. These results represent the deepest upper limits achieved by the MWA to date and provide the first evidence of the heated intergalactic medium at redshifts z = 6.5 to 7.0. 

Many mass spectrometry methods using various ionization sources provide bulk composition of airborne particles, but little is known about the surface species that play a major role in determining their physicochemical properties that impact air quality, climate, and health. The present work shows that the composition of surface layers of atmospherically relevant submicron organic particles can be probed without the use of an external ionization source. Solid dicarboxylic acid particles are used as models, with glutaric acid being the most efficient at generating ions. Coating with small diacids or products from α-pinene ozonolysis demonstrates that ions are ejected from the surface, providing surface molecular characterization of organic particles on the fly. This unique approach provides a path forward for elucidating the role of the surface in determining chemical and physical properties of particles, including heterogeneous reactions, particle growth, water uptake, and interactions with biological systems. 

ABSTRACT Interferometric experiments designed to detect the highly redshifted 21-cm signal from neutral hydrogen are producing increasingly stringent constraints on the 21-cm power spectrum, but some k-modes remain systematics-dominated. Mutual coupling is a major systematic that must be overcome in order to detect the 21-cm signal, and simulations that reproduce effects seen in the data can guide strategies for mitigating mutual coupling. In this paper, we analyse 12 nights of data from the Hydrogen Epoch of Reionization Array and compare the data against simulations that include a computationally efficient and physically motivated semi-analytic treatment of mutual coupling. We find that simulated coupling features qualitatively agree with coupling features in the data; however, coupling features in the data are brighter than the simulated features, indicating the presence of additional coupling mechanisms not captured by our model. We explore the use of fringe-rate filters as mutual coupling mitigation tools and use our simulations to investigate the effects of mutual coupling on a simulated cosmological 21-cm power spectrum in a ‘worst case’ scenario where the foregrounds are particularly bright. We find that mutual coupling contaminates a large portion of the ‘EoR Window’, and the contamination is several orders-of-magnitude larger than our simulated cosmic signal across a wide range of cosmological Fourier modes. While our fiducial fringe-rate filtering strategy reduces mutual coupling by roughly a factor of 100 in power, a non-negligible amount of coupling cannot be excised with fringe-rate filters, so more sophisticated mitigation strategies are required.