recentpopularlog in


« earlier   
Stop Using word2vec | Stitch Fix Technology – Multithreaded
When I started playing with word2vec four years ago I needed (and luckily had) tons of supercomputer time. But because of advances in our understanding of w...
ai  nlp  language  machinelearning  text  algorithms  analysis 
22 hours ago by geetarista
[1705.04058] Neural Style Transfer: A Review
The recent work of Gatys et al. demonstrated the power of Convolutional Neural Networks (CNN) in creating artistic fantastic imagery by separating and recombing the image content and style. This process of using CNN to migrate the semantic content of one image to different styles is referred to as Neural Style Transfer. Since then, Neural Style Transfer has become a trending topic both in academic literature and industrial applications. It is receiving increasing attention from computer vision researchers and several methods are proposed to either improve or extend the original neural algorithm proposed by Gatys et al. However, there is no comprehensive survey presenting and summarizing recent Neural Style Transfer literature. This review aims to provide an overview of the current progress towards Neural Style Transfer, as well as discussing its various applications and open problems for future research.
generative-art  neural-networks  convolutional-networks  algorithms  rather-interesting  nudge-targets  machine-learning 
yesterday by Vaguery
[1708.01932] Colorings beyond Fox: the other linear Alexander quandles
This article is about applications of linear algebra to knot theory. For example, for odd prime p, there is a rule (given in the article) for coloring the arcs of a knot or link diagram from the residues mod p. This is a knot invariant in the sense that if a diagram of the knot under study admits such a coloring, then so does any other diagram of the same knot. This is called p-colorability. It is also associated to systems of linear homogeneous equations over the residues mod p, by regarding the arcs of the diagram as variables and assigning the equation "twice the over-arc minus the sum of the under-arcs equals zero" to each crossing. The knot invariant is here the existence or non-existence of non-trivial solutions of these systems of equations, when working over the integers modulo p (a non-trivial solution is such that not all variables take up the same value). Another knot invariant is the minimum number of distinct colors (values) these non-trivial solutions require, should they exist. This corresponds to finding a basis, supported by a diagram, in which these solutions take up the least number of distinct values. The actual minimum is hard to calculate, in general. For the first few primes, less than 17, it depends only on the prime, p, and not on the specific knots that admit non-trivial solutions, modulo p. For primes larger than 13 this is an open problem. In this article, we begin the exploration of other generalizations of these colorings (which also involve systems of linear homogeneous equations mod p) and we give lower bounds for the number of colors.
knot-theory  topology  algorithms  representation  rather-interesting  to-understand 
yesterday by Vaguery
[1006.4176] Unknotting Unknots
A knot is an an embedding of a circle into three-dimensional space. We say that a knot is unknotted if there is an ambient isotopy of the embedding to a standard circle. By representing knots via planar diagrams, we discuss the problem of unknotting a knot diagram when we know that it is unknotted. This problem is surprisingly difficult, since it has been shown that knot diagrams may need to be made more complicated before they may be simplified. We do not yet know, however, how much more complicated they must get. We give an introduction to the work of Dynnikov who discovered the key use of arc--presentations to solve the problem of finding a way to detect the unknot directly from a diagram of the knot. Using Dynnikov's work, we show how to obtain a quadratic upper bound for the number of crossings that must be introduced into a sequence of unknotting moves. We also apply Dynnikov's results to find an upper bound for the number of moves required in an unknotting sequence.
knot-theory  rather-interesting  representation  algorithms  classification  nudge-targets  consider:looking-to-see  consider:feature-discovery 
yesterday by Vaguery
Lowdown Diffing Engine
In this paper, I briefly describe the “diff” engine used in lowdown-diff(1) tool in lowdown. The work is motivated by the need to provide formatted output describing the difference between two documents—specifically, formatted PDF via the -Tms output.

This documents an early work in progress—both source code and documentation. The source is documented fully in diff.c. This paper itself is available as, or downloadable as diff.pdf. Please direct comments to me by e-mail or just use the GitHub interface.

For a quick example of this functionality, see diff.diff.html, which shows the difference between this document and a [fabricated] earlier version.
algorithms  tools  markdown 
yesterday by micktwomey
'Sandwiching': How to outsmart Instagram’s algorithm to drive engagement | Glossy
Kronengold and his team found that, while the most effective posting cadence and strategy varies by brand, for the most part, posting a promotional post the day after a well-performing post (with a high number of likes and comments), and then following it up the subsequent day with another highly engaging post, yields the best results for the promoted content.
instagram  algorithms  how-to 
yesterday by dancall
The MICrONS program aims to take a quantum leap in by reverse-engineering of the brain
algorithms  MachineLearning  from twitter_favs
2 days ago by sclopit

Copy this bookmark:

to read