Unsupervised deep learning shows that the brain disentangles faces into semantically meaningful factors, like age or the presence of a smile, at the single neuron level @irinavlh @doristsao @demishassabis @DeepMind @Caltech Tweet added by Nature Communications @NatureComms

Nature Communications

3 years

Unsupervised deep learning shows that the brain disentangles faces into semantically meaningful factors, like age or the presence of a smile, at the single neuron level @irinavlh @doristsao @demishassabis @DeepMind @Caltech

7

158

661

Himanshu

@plasticwatercan

3 years

@NatureComms @irinavlh @doristsao @demishassabis @DeepMind @Caltech Can someone ELI5 this?

3

0

3

Rudolph Ukka

@RudolphUkka

3 years

@NatureComms @DeepMind @irinavlh @doristsao @demishassabis @Caltech That's a very clever neuron!

0

2

Harrison Kennedy

@harrisonkennedy

3 years

@NatureComms @irinavlh @doristsao @demishassabis @DeepMind @Caltech @ScottAdamsSays simplified processing

0

1

Sujay Parekh

@sujay_parekh

3 years

@NatureComms @irinavlh @doristsao @demishassabis @DeepMind @Caltech Interesting that macaque neurons have coded for human traits.

0

Sandeep

@sandeepvbelure

3 years

@NatureComms @irinavlh @doristsao @demishassabis @DeepMind @Caltech But is there a hierarchy among the factors these neurons recognise, for speed and accuracy of recognition?

Sandeep

@sandeepvbelure

3 years

For AGI to occur, robots would need to be conscious. They would probably need to identify themselves in a mirror. Is there a way to build an intuitive face identifier, one that does not use distances b/w face landmarks?

0

Replies