Gaussian process state-space models are a widely used modelling paradigm for learning and estimation in dynamical systems. Reduced-rank Gaussian process state-space models combine spectral characterization of dynamical systems with Hilbert space methods to enable learning, which scales linearly with the length of the time series. However, the current state of the art algorithms struggle to deal efficiently with the dimensionality of the state-space parameters. In this work, we propose a domain-aware reduced-rank Gaussian process model which uses the relationship between state dimensions to reduce the computational cost, by potentially orders of magnitude in comparison to the state-of-the-art, while maintaining comparable performance. We conduct simulations to showcase the performance and reduced computational cost of our proposed method, in comparison with the state-of-the-art learning method, and propose future research directions. @en

Low-dimensional hybrid perovskites have emerged as promising materials for optoelectronic applications. Although these materials have already demonstrated enhanced stability as compared to their three-dimensional perovskite analogues, their functionality has been limited by the insulating character of the organic moieties that primarily play a structure-directing role. This is particularly the case for the layered (2D) perovskite materials based on formamidinium lead iodide (FAPbI3) that remain scarce. We demonstrate a low-dimensional hybrid perovskite material based on a SPbI4 composition incorporating an electroactive naphthalenediimide (NDI) moiety as an organic spacer (S) between the perovskite slabs and evidence the propensity of the spacer to stabilize the α-FAPbI3 perovskite phase in hybrid low-dimensional SFAn-1PbnI3n+1 perovskite compositions. This has been investigated by means of solid-state nuclear magnetic resonance spectroscopy in conjunction with molecular dynamics simulations and density functional theory calculations. Theoretical calculations suggest an electronic contribution of the organic spacer to the resulting optoelectronic properties, which was confirmed by transient absorption spectroscopy. We have further analyzed these materials by time-resolved microwave conductivity measurements, revealing challenges for their application in photovoltaics.

@en