CGSC 1001 Lecture Notes - Recommender System, Netflix, Distributed Computing
Document Summary
Disembodied software: how netflix works, cognitive system is the amazon recommender system, we ask it to do something, and therefore user input. Human memory of fact & perception: sensory memory: is like a scratch pad (for vision), or a bit or recording tape (for hearing), that can be overwritten with new things. It is rewritten every few seconds by new perceptions: short-term memory: is a temporary store. Some things from it end up becoming long term memories: long-term memory: probably stored here forever, though we might have trouble with retrieval. Sensory short term long term memory. Computer memory: hash tables, schemata, frames, scripts, logical sentences, disk, flash memory. Basic building of the computer: bitmaps, activation patterns, weights in links. Memory in distributed systems examples: books, brains, fingers, notes, arrows and signs. Learning: learning is changing memory with the purpose of preparing a system for better action in the future.