This strong emphasis on 'learning' is evident across the AI community. Can see it in the difference between arXiv and bioRxiv papers about intelligence. http://mdcrosby.com/blog/artificialvsbiological.html …
D’oh. @DeepMindAI is at it again. Cool new model of theory of mind, but same old misleading emphasis of learning at the expense of innateness.
I discuss, briefly, in a new piece @medium.https://medium.com/@GaryMarcus/deepminds-misleading-campaign-against-innateness-a2ea6eb4d0ba?source=twitterShare-d7e74ac84d28-1519573619 …
-
-
- 2 more replies
New conversation -
-
-
I have yet to hear the 'principled' way that you would derive these innate cognitive machinery. This is analogous to demanding that physics has a 'principled' way of deriving the laws of physics. The principled way is through doing experiments.
-
What are the boundaries of your argument of innate machinery? Do researchers have to derive innate machinery from the laws of classical physics or quantum mechanics?
- 7 more replies
New conversation -
-
-
You might find
@vicariousai approach to prior structure called "scaffolding" novel in prior structure. Q: Do Captchas require "generalization to distributions that are not represented in the training set" or is the problem actually low dimensional? https://www.vicarious.com/2017/10/26/common-sense-cortex-and-captcha/ …Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Hmm, good points.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
From conception to cells dividing to massive cell population moving to specialization to a growing human and on to learning and adapting to a growing population and its complexities. Machines are a human-made extension of humans, nothing less. http://abstractionphysics.net
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.