23 JAN 2015 by ideonexus

 Morphological Diversity of Random Town

Tally spent the day walking around the city, marveling at how different it was from her own. She saw new pretties and uglies hanging out together, friends that the operation hadn't separated. And littlies clinging to their ugly older brothers and sisters instead of being stuck in Crumblyville with their parents. Those small changes were almost as surprising as the wild facial structures, skin textures, and body mods she encountered. Almost. It might take a while to get used to coats of downy...
Folksonomies: science fiction
Folksonomies: science fiction
  1  notes
 
21 APR 2014 by ideonexus

 Perceptrons and Sigmoid Neurons

A way you can think about the perceptron is that it's a device that makes decisions by weighing up evidence. Let me give an example. It's not a very realistic example, but it's easy to understand, and we'll soon get to more realistic examples. Suppose the weekend is coming up, and you've heard that there's going to be a cheese festival in your city. You like cheese, and are trying to decide whether or not to go to the festival. You might make your decision by weighing up three factors: 1. Is...
  1  notes

Two tool for machine learning.

29 DEC 2013 by ideonexus

 Douglas Engelbart's Idea of Small Changes

I'm reminded of Douglas Engelbart's classic paper "Augmenting Human itellect,"2 on his belief in the power of computers. He wrote this in 1962, way before the PC, and argued that it's better to improve and facilitate the tiny things we do every day than it is to attempt to replace entire human jobs with monolithic machines. A novel-writing machine, if one were invented, just automates the process of writing novels, and it's limited to novels. But making a small improvement to a pencil, for ex...
Folksonomies: invention change
Folksonomies: invention change
  1  notes

Make a change to novel-writing and you've affected a small, specific domain, but improve the pencil, and you've impacted a wide range of domains.

20 SEP 2011 by ideonexus

 Evolution Takes Time

True, breeders haven’t turned a cat into a dog, and laboratory studies haven’t turned a bacterium into an amoeba (although, as we’ve seen, new bacterial species have arisen in the lab). But it is foolish to think that these are serious objections to natural selection. Big transformations take time—huge spans of it. To really see the power of selection, we must extrapolate the small changes that selection creates in our lifetime over the millions of years that it has really had to work...
Folksonomies: evolution time
Folksonomies: evolution time
  1  notes

A river formed the Grand Canyon, so we know that small processes can have huge effects given enough time.