Isn't the whole point of AGI to build a synthetic version of something we've observed in the real world? I agree progress should be tracked by something other than "well it looks like a brain!" - but biological plausibility should still be considered as a kind of progress, right?
This tweet is brilliant; required reading for anyone who cares about biological plausibility.https://twitter.com/vdv7/status/949702899810095104 …
-
-
-
At the end of the day - we *are* trying to copy what nature has already achieved - intelligence.
- 7 more replies
New conversation -
-
-
Mechanism of learning is probably more important for AGI than the underlying representation.
-
The idea of learning from data is itself problematic for AGI, given that many critical human activities aren’t described by it - even knowledge-creating ones like pondering on a long walk.
End of conversation
New conversation -
-
-
This is brilliantly worded. Machine learning should only take light inspiration from neuroscience, mostly because we don’t understand enough of it.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Agree that apeing (Ho Ho) biological intelligence is not the end in itself but is is not likely/ obvious that AI researchers can learn from the many highly capable BIs that exist? Plane/bird analogy is instructive, not a cliff to push one's opponents over.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
There is a crucial lesson we can learn from the brain. All sensory stimuli are converted into precisely timed spikes. Problem then becomes strictly temporal and a single solution works for all sensory modalities. No need to have multiple special solutions. https://sciencealert.com/case-study-vision-patient-with-bilateral-occipital-lobe-v1-injury …
-
IOW, there is no need to have different mechanisms for processing auditory, visual and tactile inputs. This is why I reject Geoffrey Hinton's capsules or Jeff Hawkins' location signal. Besides, a single mechanism makes it easy to draw analogies between disparate concepts.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.