You're going to have to explain this better.....I'm not 'seein it'.....Look how easy I can explain image recognition - convert an image into lines using a sobel line filter, predict the next pixels, but with a distance pattern recognizer i.e. you know the memory ex. 'boat' and see 'b--o--?' and so you predict the a and t with the same error in distance, this also allows recognition of varying sizes, brightnesses etc, cux it all is bigger/ brighter / etc.
What about a compressed file sitting in enwik9, so you take it and uncompress it and then compress it better but have to know to compress it back to a lousy form using their (?) algorithm. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Td08f1cb9cdd5e5d9-M72a21cc9aef2bf9b378d3170 Delivery options: https://agi.topicbox.com/groups/agi/subscription
