Weifeng Li — Page 2

arXiv2014

A Gapless MoS$_2$ Allotrope Possessing Both Massless Dirac and Heavy Fermions

MoS$_2$, a member of transition metal dichalcogenides (TMDs), recently emerged as one of the fastest growing two-dimensional materials due to its fascinating mechanical, thermal, electronic and optical properties. Unlike graphene which possesses massless Dirac fermions with ultra-high electron mobility, monolayer MoS$_2$ is a direct band gap semiconductor. An interesting question arises: Can monolayer MoS$_2$ also possess massless Dirac fermions with ultra-high electron mobility? Here, using first-principles calculations, we show that a monolayer MoS$_2$ allotrope, which consists of repeated square-octagon rings (abbreviated as so-MoS$_2$ to distinguish from the normal hexagonal lattice, h-MoS$_2$) possesses both massless Dirac fermions and heavy fermions. Distinct from the $p$-orbital Dirac fermions of graphene, the Dirac fermions of so-MoS$_2$ are $d$-electrons and possess Fermi velocity comparable to that of graphene. The Dirac cone structure in so-MoS$_2$ demonstrated here greatly enriches our understanding on the physical properties of TMDs and opens up new possibilities for developing novel electronic/spintronic devices.

cond-mat.mes-hall↗

arXiv2013

Feature Learning with Gaussian Restricted Boltzmann Machine for Robust Speech Recognition

In this paper, we first present a new variant of Gaussian restricted Boltzmann machine (GRBM) called multivariate Gaussian restricted Boltzmann machine (MGRBM), with its definition and learning algorithm. Then we propose using a learned GRBM or MGRBM to extract better features for robust speech recognition. Our experiments on Aurora2 show that both GRBM-extracted and MGRBM-extracted feature performs much better than Mel-frequency cepstral coefficient (MFCC) with either HMM-GMM or hybrid HMM-deep neural network (DNN) acoustic model, and MGRBM-extracted feature is slightly better.

cs.CL↗