DrEliza was even capable of finding new cures for conditions that were NOT coded in it's knowledge base, but it was too slow due to combinatorial explosion to take over medicine.
Then, I patented an incredibly fast way of doing the AI, but getting from the old text-based programming paradigm to the new multiple-stack pointer-based system would involve starting over from scratch. Attempts to fund a startup ran aground, mostly because money people couldn't wrap their minds around stacks and pointers. Might this fit what you are looking for? Steve On Sat, Nov 16, 2019, 2:47 PM <immortal.discover...@gmail.com> wrote: > Open AI is implementing Unsupervised Learning though: > https://openai.com/blog/better-language-models/ > Unsupervised Learning is auto sorting, you have a compression/error goal > still but that's it. > Supervised Learning is when you give it many labels and tell it to match > some data or gather new data. Imitation Learning. > Reinforced Learning is when you get reward for randomized outputs or > randomized inputs. You know what decisions you have and decide at each > step, with probabilities. > > Unsupervised Learning (training) is best but hardest. Search is using > Self-Attention and digging through large search space *after* the last > training update. > *Artificial General Intelligence List <https://agi.topicbox.com/latest>* > / AGI / see discussions <https://agi.topicbox.com/groups/agi> + > participants <https://agi.topicbox.com/groups/agi/members> + delivery > options <https://agi.topicbox.com/groups/agi/subscription> Permalink > <https://agi.topicbox.com/groups/agi/T26a5f8008aa0b4f8-M7ed673e233572839f7906605> > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T26a5f8008aa0b4f8-M875660e6351a512f99571928 Delivery options: https://agi.topicbox.com/groups/agi/subscription
