Mohammad Javadi | Searcharxiv

arXiv2020

Kinetic Monte Carlo Simulation of Single-Electron Multiple-Trapping Transport in Disordered Media

The conventional single-particle Monte Carlo simulation of charge transport in disordered media is based on the truncated density of localized states (DOLS) which benefits from very short time execution. Although this model successfully clarifies the properties of electron transport in moderately disordered media, it overestimates the electron diffusion coefficient for strongly disordered media. The origin of this deviation is discussed in terms of zero-temperature approximation in the truncated DOLS and the ignorance of spatial occupation of localized states. Here, based on the multiple-trapping regime we introduce a modified single-particle kinetic Monte Carlo model that can be used to investigate the electron transport in any disordered media independent from the value of disorder parameter. In the proposed model, instead of using a truncated DOLS we imply the raw DOLS. In addition, we have introduced an occupation index for localized states to consider the effect of spatial occupation of trap sites. The proposed model is justified in a simple cubic lattice of trap sites for broad interval of disorder parameters, Fermi levels, and temperatures.

cond-mat.dis-nn↗

arXiv2018

Zoom-RNN: A Novel Method for Person Recognition Using Recurrent Neural Networks

The overwhelming popularity of social media has resulted in bulk amounts of personal photos being uploaded to the internet every day. Since these photos are taken in unconstrained settings, recognizing the identities of people among the photos remains a challenge. Studies have indicated that utilizing evidence other than face appearance improves the performance of person recognition systems. In this work, we aim to take advantage of additional cues obtained from different body regions in a zooming in fashion for person recognition. Hence, we present Zoom-RNN, a novel method based on recurrent neural networks for combining evidence extracted from the whole body, upper body, and head regions. Our model is evaluated on a challenging dataset, namely People In Photo Albums (PIPA), and we demonstrate that employing our system improves the performance of conventional fusion methods by a noticeable margin.

cs.CV↗