output
20182025
most citedDropout is a special case of the stochastic delta rule: faster and more accurate deep learning

8 citations

5 papers

physics.atom-ph2025★ 3 cited

Single-Beam Magneto-Optical Trap in Back-to-Back Pyramidal and Conical Mirrors

Timothy H. Nguyen, Mariam Mchedlidze, Guanghui Su +3

A three-dimensional magneto-optical trap (MOT), as an efficient method of producing cold atoms from room-temperature atomic vapor, has been widely used to develop atomic sensors. V…

math.GT2024

Exotic aspherical 4-manifolds

Michael Davis, Kyle Hayden, Jingyin Huang +2

We construct closed, aspherical, smooth 4-manifolds that are homeomorphic but not diffeomorphic. These provide counterexamples to a smooth analog of the Borel conjecture in dimensi…

math.GR2024

Low complexity among principal fully irreducible elements of Out()

Naomi Andrew, Paige Hillen, Robert Alonzo Lyman +1

We find the shortest realized stretch factor for a fully irreducible and show that it is realized by a "principal" fully irreducible element. We also show t…

math.OC2018

Ameso Optimization: a Relaxation of Discrete Midpoint Convexity

Wen Chen, Odysseas Kanavetas, Michael N. Katehakis

In this paper we introduce the Ameso optimization problem, a special class of discrete optimization problems. We establish its basic properties and investigate the relation between…

cs.LG2018★ 8 cited

Dropout is a special case of the stochastic delta rule: faster and more accurate deep learning

Noah Frazier-Logue, Stephen José Hanson

Multi-layer neural networks have lead to remarkable performance on many kinds of benchmark tasks in text, speech and image processing. Nonlinear parameter estimation in hierarchica…