Category: Front Page

Journey through the layers of the mind

first tests playing with #deepdream #inceptionism

A visualization of what’s happening inside the mind of an artificial neural network.

By recognizing forms in these images, your mind is already reflecting what’s going on in the software, projecting its own bias onto what it sees. You think you are seeing things, perhaps puppies, slugs, birds, reptiles etc. If you look carefully, that’s not what’s in there. But those are the closest things your mind can match to what it’s seeing. Your mind is struggling to put together images based on what you know. And that’s exactly what’s happening in the software. And you’ve been training your mind for years, probably decades. These neural networks are usually trained for a few hours, days or weeks.


In non-technical speak:

An artificial neural network can be thought of as analogous to a brain (immensely, immensely, immensely simplified. nothing like a brain really). It consists of layers of neurons and connections between neurons. Information is stored in this network as ‘weights’ (strengths) of connections between neurons. Low layers (i.e. closer to the input, e.g. ‘eyes’) store (and recognise) low level abstract features (corners, edges, orientations etc.) and higher layers store (and recognise) higher level features. This is analogous to how information is stored in the mammalian cerebral cortex (e.g. our brain).

Here a neural network has been ‘trained’ on millions of images – i.e. the images have been fed into the network, and the network has ‘learnt’ about them (establishes weights / strengths for each neuron). (NB. This is a specific database of images fed into the network known as ImageNet )

Then when the network is fed a new unknown image (e.g. me), it tries to make sense of (i.e. recognise) this new image in context of what it already knows, i.e. what it’s already been trained on.

This can be thought of as asking the network “Based on what you’ve seen / what you know, what do you think this is?”, and is analogous to you recognising objects in clouds or ink / rorschach tests etc.

The effect is further exaggerated by encouraging the algorithm to generate an image of what it ‘thinks’ it is seeing, and feeding that image back into the input. Then it’s asked to reevaluate, creating a positive feedback loop, reinforcing the biased misinterpretation.

This is like asking you to draw what you think you see in the clouds, and then asking you to look at your drawing and draw what you think you are seeing in your drawing etc,

That last sentence was actually not fully accurate. It would be accurate, if instead of asking you to draw what you think you saw in the clouds, we scanned your brain, looked at a particular group of neurons, reconstructed an image based on the firing patterns of those neurons, based on the in-between representational states in your brain, and gave *that* image to you to look at. Then you would try to make sense of (i.e. recognise) *that* image, and the whole process will be repeated.

We aren’t actually asking the system what it thinks the image is, we’re extracting the image from somewhere inside the network. From any one of the layers. Since different layers store different levels of abstraction and detail, picking different layers to generate the ‘internal picture’ hi-lights different features.

All based on the google research by Alexander Mordvintsev, Software Engineer, Christopher Olah, Software Engineering Intern and Mike Tyka, Software Engineer

(View on Vimeo)

Colossus by Pat Vale

follow me on Instagram –
follow me on facebook –

Massive thanks to John Barber for the incredible musical score. Hear more of his work at and

Thanks to Colorist Daniel Silverman at MPC (and Ariella Amrami)

Dom del Torto at Big Animal

Colossus is a drawing that I made in New York during December 2014.

(View on Vimeo)

The Fallen of World War II

An animated data-driven documentary about war and peace, The Fallen of World War II looks at the human cost of the second World War and sizes up the numbers to other wars in history, including trends in recent conflicts.

Visit for the interactive version and more information

Written, directed, coded, narrated by
Sound and music by

(View on Vimeo)

A portion of a ‘body density map’ created in 1918 by…

A portion of a ‘body density map’ created in 1918 by Lieutenant Colonel Arthur Messer. Lt.-Col. Messer was engaged in an attempt to record battlefield crosses and have the bodies of soldiers moved and re-interred in cemeteries.

This is a section of a map depicting an area of the Somme battlefield, one of the largest battles of the war. Each small square on this map is an area of 83 by 83 yards; each blue number written in these squares represents the number of soldiers killed in that area. So, for example, in the top-right quadrant of area S-10, Lt.-Col. Messer counted 808 dead.

The Battle of the Somme lasted for almost three months in the summer of 1916. More than 1,000,000 men were wounded or killed; it was one of the bloodiest battles in human history.

(View on Tumblr)

Lunar economic zone by Zhan Wang

See more architecture and design movies at

Architectural Association graduate Zhan Wang has produced an animation depicting a fictional technotopian future scenario in which China has built a giant port to distribute minerals mined from the moon.

Zhan Wang’s Lunar Economic Zone project imagines a celebration taking place in Shenzhen in the year 2028 to mark the arrival of the first shipment of lunar minerals.

The animation portrays the architecture and infrastructure required by such a system and the way the parade might be propagandised to present China’s technological and economic prowess through the lens of the global media.

See the full story on Dezeen:

(View on Vimeo)

xplanes: Apparently I have posted only eight times in the last…


Apparently I have posted only eight times in the last year. This is due to a new addition to the family, who is now too big for his skeleton suit.

All I have for you at the moment is this: the spooky tag

(above image: ““Farman 40 of pilot Lt. Jaumotte and observer sLt. Wouters”, from a website/source that no longer exists!)

Hope to be “back” soon.

(View on Tumblr)