The thousand faces of CaffeNet

Here are the thousand image categories in the ImageNet Challenge data visualisation context, rendered onto a webcam photo of me using the deepdraw technique.

This is the set of image classifications which the standard Deepdreams imagery is based on: as more than one hundred of the categories are dog breeds …

more ...

The Homunculus

This is the first in a series of posts about the way we think about programming, the influence these ideas have on how we write code and what languages we choose, and why I like the functional programming paradigm in general and Haskell in particular.

Consumer notification: I've worked as …

more ...

Items from the ImageNet Large Scale Image Recognition Challenge which I'm pretty sure I wouldn't be able to identify

n01440764 tench, Tinca tinca
n01496331 electric ray, crampfish, numbfish, torpedo
n01530575 brambling, Fringilla montifringilla
n01532829 house finch, linnet, Carpodacus mexicanus
n01534433 junco, snowbird
n01537544 indigo bunting, indigo finch, indigo bird, Passerina cyanea
n01601694 water ouzel, dipper
n01608432 kite
n01622779 great grey owl, great gray owl, Strix nebulosa
n01629819 European fire …
more ...

Lucid dreaming with neural nets

The Deepdreams algorithm takes a deep neural net trained for image-recognition and uses a forced-feedback loop to induce a kind of machine-pareidolia or hallucination, with beautiful and often nightmarish results:

Cassowary

The original algorithm picks details from all of the neural net's training data, based on resemblances to the original image …

more ...