On Saturday, January 04, 2020, at 1:37 AM, rouncer81 wrote: > I dont think a cell would ever self ignite, wouldn't it always be a pattern > off the sensor. > How is something supposed to self ignite, it goes against determinism it > seems, so it could be out-ruled! > Do you know of another architecture other than Transformers or mine that is not only going to perform as well as GPT-2 but also is simple and simple to understand and you can actually explain it to me so I can start at it already? Oh, no and no! GPT-2 is the best yet and is a monster to understand clear, no one here can explain it clear that they won't even try. My goal is to get the same results using this simple architecture. Once I do that then I have a more realistic architecture.
On Saturday, January 04, 2020, at 1:37 AM, rouncer81 wrote: > Why try to be the full natural brain, when simple artificial computer brainz > work well (even tho u might not know that yet.) and your not imposing on > nature so much, and itll still cause quite a stir even if the robot isnt as > capable as a human in all respects, you still get all the perks of > artificial creations. (all the superpowers you get from being artificial and > man made!!!) > I wake up every morning hearing/seeing 'AGI' 'computer' features. It probably is energy efficient to just let external stimuli easily leak into these nodes so power them using free energy. They are accessed so much the channels are wide open. And most features link to them because you think about it all day. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T409fc28ec41e6e3a-M889b6a49d3565b8424b384c1 Delivery options: https://agi.topicbox.com/groups/agi/subscription
