arXiv subjects

Haonan Huang

Publications and source records attributed to Haonan Huang.

24 records · Page 2Linked to original sources

arXiv2022

Multi-view Data Classification with a Label-driven Auto-weighted Strategy

Distinguishing the importance of views has proven to be quite helpful for semi-supervised multi-view learning models. However, existing strategies cannot take advantage of semi-supervised information, only distinguishing the importance of views from a data feature perspective, which is often influenced by low-quality views then leading to poor performance. In this paper, by establishing a link between labeled data and the importance of different views, we propose an auto-weighted strategy to evaluate the importance of views from a label perspective to avoid the negative impact of unimportant or low-quality views. Based on this strategy, we propose a transductive semi-supervised auto-weighted multi-view classification model. The initialization of the proposed model can be effectively determined by labeled data, which is practical. The model is decoupled into three small-scale sub-problems that can efficiently be optimized with a local convergence guarantee. The experimental results on classification tasks show that the proposed method achieves optimal or sub-optimal classification accuracy at the lowest computational cost compared to other related methods, and the weight change experiments show that our proposed strategy can distinguish view importance more accurately than other related strategies on multi-view datasets with low-quality views.

cs.CV↗

arXiv2020

Spin-dependent tunneling between individual superconducting bound states

Magnetic impurities on superconductors induce discrete bound levels inside the superconducting gap, known as Yu-Shiba-Rusinov (YSR) states. YSR levels are fully spin-polarized such that the tunneling between YSR states depends on their relative spin orientation. Here, we use scanning tunneling spectroscopy to resolve the spin dynamics in the tunneling process between two YSR states by experimentally extracting the angle between the spins. To this end, we exploit the ratio of thermally activated and direct spectral features in the measurement to directly extract the relative spin orientation between the two YSR states. We find freely rotating spins down to 7mK, indicating a purely paramagnetic nature of the impurities. Such a non-collinear spin alignment is essential not only for producing Majorana bound states but also as an outlook manipulating and moving the Majorana state onto the tip.

cond-mat.supr-con↗

arXiv2020

Partially Shared Semi-supervised Deep Matrix Factorization with Multi-view Data

Since many real-world data can be described from multiple views, multi-view learning has attracted considerable attention. Various methods have been proposed and successfully applied to multi-view learning, typically based on matrix factorization models. Recently, it is extended to the deep structure to exploit the hierarchical information of multi-view data, but the view-specific features and the label information are seldom considered. To address these concerns, we present a partially shared semi-supervised deep matrix factorization model (PSDMF). By integrating the partially shared deep decomposition structure, graph regularization and the semi-supervised regression model, PSDMF can learn a compact and discriminative representation through eliminating the effects of uncorrelated information. In addition, we develop an efficient iterative updating algorithm for PSDMF. Extensive experiments on five benchmark datasets demonstrate that PSDMF can achieve better performance than the state-of-the-art multi-view learning approaches. The MATLAB source code is available at https://github.com/libertyhhn/PartiallySharedDMF.

cs.LG↗

arXiv2020

Interplay Between Yu-Shiba-Rusinov States and Multiple Andreev Reflections

Motivated by recent scanning tunneling microscopy experiments on single magnetic impurities on superconducting surfaces, we present here a comprehensive theoretical study of the interplay between Yu-Shiba-Rusinov bound states and (multiple) Andreev reflections. Our theory is based on a combination of an Anderson model with broken spin degeneracy and nonequilibrium Green's function techniques that allows us to describe the electronic transport through a magnetic impurity coupled to superconducting leads for arbitrary junction transparency. Using this combination we are able to elucidate the different tunneling processes that give a significant contribution to the subgap transport. In particular, we predict the occurrence of a large variety of Andreev reflections mediated by Yu-Shiba-Rusinov bound states that clearly differ from the standard Andreev processes in non-magnetic systems. Moreover, we provide concrete guidelines on how to experimentally identify the subgap features originating from these tunneling events. Overall, our work provides new insight into the role of the spin degree of freedom in Andreev transport physics.

cond-mat.supr-con↗

arXiv2019

Tunneling dynamics between superconducting bound states at the atomic limit

Despite plenty of room at the bottom, there is a limit to the miniaturization of every process. For charge transport this is realized by the coupling of single discrete energy levels at the atomic scale. Here, we demonstrate sequential tunneling between parity protected Yu-Shiba-Rusinov (YSR) states bound to magnetic impurities located on the superconducting tip and sample of a scanning tunneling microscope at 10 mK. We reduce the relaxation of the excited YSR state to the bare minimum and find an enhanced lifetime for single quasiparticle levels. Our work offers a way to characterize and to manipulate coupled superconducting bound states, such as Andreev levels, YSR states, or Majorana bound states.

cond-mat.supr-con↗

arXiv2019

Magnetic Impurities on Superconducting Surfaces: Phase Transitions and the Role of Impurity-Substrate Hybridization

Spin-dependent scattering from magnetic impurities inside a superconductor gives rise to Yu-Shiba-Rusinov (YSR) states within the superconducting gap. As such, YSR states have been very successfully modeled with an effective scattering potential (Kondo impurity model). Using a scanning tunneling microscope, we exploit the proximity of the tip to a surface impurity and its influence on the YSR state to make a quantitative connection between the YSR state energy and the impurity-substrate hybridization. We corroborate the coupling between impurity and substrate as a key energy scale for surface derived YSR states using the Anderson impurity model in the mean field approximation, which accurately explains our observations. The model allows to decide on which side of the quantum phase transition the system resides based on additional conductance measurements. We propose that the Anderson impurity model is much more appropriate to describe YSR states from impurities on a superconducting surface than the Kondo impurity model, which is more appropriate for impurities inside a superconductor. We thus provide a first step towards a more quantitative comparison of experimental data with fully correlated calculations based on the Anderson impurity model.

cond-mat.supr-con↗