If you just have a single problem to solve then fine go ahead and use a neural network. But if you want to do science and understand how to choose architectures or how to go to a new problem you have to understand what different architectures can and cannot do.
Kubrick's vision seemed to be that humans are doomed whereas Clarke's is...
We wanted to solve robot problems and needed some vision action reasoning...
I think Lenat is headed in the right direction but someone needs to include...
When David Marr at MIT moved into computer vision he generated a lot of...
There was a failure to recognize the deep problems in AI for instance those...
Societies need rules that make no sense for individuals. For example it...
Whenever you deal with science fiction you are setting up a world of rules. I...
I sort of was good at writing essays. I was never very good at mathematics...
Science fiction is no more written for scientists that ghost stories are...
We must always emphasize research and development of science and mathematics...
The relevance of Marxism to science is that it removes it from its imagined...
The recognition of the art that informs all pure science need not mean the...
It is characteristic of science that the full explanations are often seized...
A third ideal that has made its way in the modern world is reliance on reason...
I do love science fiction but it's not really a genre unto itself it always...
If you look at the purported dangers of salt or fat there is no consensus of...