Hi I’ve been paying attention and thought I’d throw my own pennies worth into the mix.
The problem with any human derived idea/ theorem like ‘no free lunch’ is that… it’s formulated/ constructed by humans, and as such can only be applied using the depth/ expanse of our own knowledge, understanding and experience. I’ve always considered this a very narrow, almost arrogant perspective. Even mathematical proof is relative to our current understanding of our reality. We should use our theories/ knowledge as a guide only, not set rules. @Danko >Otherwise, we would have only one machine learning algorithm, not multiple ones. Each of them works better in one situation and worse in another. We only know of one true intelligent system, the multiple AI algorithms you speak of have arose because they each only tap one aspect the human connectomes schema, this is obviously why they are all based on similar model (nodes/ connections). >Then we go and say "let me make a network of neurons as similar to human brain as possible". Maybe this works. The NFL theorems, goes again telling us "Nope, included that too in my proof. Not working." But it does indeed work; what you are stating is that no one ‘so far’ had figured it out, you are again applying a narrow/ relative rule NFL which is limited/ constrained by human/ your understanding/ knowledge. Even E=MC2 should be taken as guide, not a rule… we as a race do not know everything. https://www.youtube.com/user/korrelan K:) ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T353f2000d499d93b-M096dd58cfd94f7cc70a0592d Delivery options: https://agi.topicbox.com/groups/agi/subscription
