output
20032026
most citedDeep Learning using Rectified Linear Units (ReLU)

2.5k citations

Showing 2020Show all

9 papers · 1 filter

cs.CL2020

Exploiting News Article Structure for Automatic Corpus Generation of Entailment Datasets

Jan Christian Blaise Cruz, Jose Kristian Resabal, James Lin +2

Transformers represent the state-of-the-art in Natural Language Processing (NLP) in recent years, proving effective even in tasks done in low-resource languages. While pretrained t…

cond-mat.mes-hall2020★ 12 cited

Guided modes and terahertz transitions for two-dimensional Dirac fermions in a smooth double-well potential

R. R. Hartmann, M. E. Portnoi

The double-well problem for the two-dimensional Dirac equation is solved for a family of quasi-one-dimensional potentials in terms of confluent Heun functions. We demonstrate that…

cond-mat.mes-hall2020★ 16 cited

Bipolar electron waveguides in graphene

R. R. Hartmann, M. E. Portnoi

We show analytically that the ability of Dirac materials to localize an electron in both a barrier and a well can be utilized to open a pseudo-gap in graphene's spectrum. By using…

cs.LG2020

Improving k-Means Clustering Performance with Disentangled Internal Representations

Abien Fred Agarap, Arnulfo P. Azcarraga

Deep clustering algorithms combine representation learning and clustering by jointly optimizing a clustering loss and a non-clustering loss. In such methods, a deep neural network…

math.DS2020★ 5 cited

Absolute concentration robustness in power law kinetic systems

Noel T. Fortun, Eduardo R. Mendoza

Absolute concentration robustness (ACR) is a condition wherein a species in a chemical kinetic system possesses the same value for any positive steady state the network may admit r…

gr-qc2020★ 44 cited

Rotating dirty black hole and its shadow

Reggie C. Pantig, Emmanuel T. Rodulfo

In this paper, we examine the effect of dark matter to a Kerr black hole of mass . The metric is derived using the Newman-Janis algorithm, where the seed metric originates from…