Deep learning AI "autoencodes" Blade Runner, recreates it so faithfully it gets a takedown notice

[Read the post]

1 Like

dude, the first rule of AI is you never show them blade runner, terminator, or short circuitā€¦

35 Likes

Donā€™t forget Small Wonder.

15 Likes

ā€¦ reduces complex information to a small subset that the neural net believes to be most significant.

Compression. Thatā€™s the correct word.

9 Likes

So AIs can dream. One step closer to consciousness.

3 Likes

ā€¦of electric sheep.

21 Likes

touche!

Doesnā€™t YouTubeā€™s content id system use a neural network to categorize videos? Wouldnā€™t that make it possible the two bits of software are actually seeing the same thing?

1 Like

ā€¦or of cuddly human babies. We shall see.

not sure but at least one of the videos contains the audio from the movieā€¦ which probably made it easy to flag.

the vimeo video seems to have no sound, but then again ā€œblade runnerā€ is in the title which may have weighted the automatic analysis.

1 Like

it was a set up, and you too have fallen prey to my evil plan.

1 Like

First of the day, wonā€™t be the last thoughā€¦

2 Likes

Well now I havenā€™t ā€“ thanx for nothinā€™

7 Likes

Is this to be an empathy test?

3 Likes

No, noā€¦ the rule on that is you never show it to any intelligence, artificial or otherwise.

6 Likes

Sheā€™s a small wonder, lovely and bright with soft curls.
Sheā€™s a small wonder, a child unlike other girls.
Sheā€™s a miracle, and I grant you
Sheā€™ll enchant you at her sight.
Sheā€™s a small wonder, and sheā€™ll make your heart take flight.

Sheā€™s fantastic, made of plastic.
Microchips here and there.
Sheā€™s a small wonder, brings love and laughter everywhere

:smiling_imp:

5 Likes

The AI autoencoded version is like viewing Blade Runner through a scanner, darkly.

17 Likes

Thereā€™s this: http://news.berkeley.edu/2011/09/22/brain-movies/

ā€œImagine tapping into the mind of a coma patient, or watching oneā€™s own
dream on YouTube. With a cutting-edge blend of brain imaging and
computer simulation, scientists at the University of California,
Berkeley, are bringing these futuristic scenarios within reach.ā€

Aw man, why didnā€™t he use the Directorā€™s Cut without the narrative dubbing?

2 Likes

The ļ¬nal convolutional layer in the encoder network Enc is reshaped (ļ¬‚at-tened) from a 4th-order Tensor of size 16x9x640x12 to a linear (ignoringthe batch depth) 2nd-order Tensor of size 92160x12. The ļ¬gure 92160 being the sum 16Ɨ9Ɨ640. The ļ¬nal layer in the encoder network Enc is fully connected to the ļ¬‚attened convolutional layer. The ļ¬nal layer is a 2nd-orderTensor of size 200x12, 200 being the dimension of latent variables z.

So no, thatā€™s not a 200-digit number. Itā€™s 200x12x(16.5 bits), which is about 12,000 decimal digits of information.

I think.

4 Likes