3 ms·
To put this into game theory lingo, I think this is because the “branching factor” for any kind of research or exploit is extremely large. So looking backwards
by janalsncm 3mo ago
To put this into game theory lingo, I think this is because the “branching factor” for any kind of research or exploit is extremely large. So looking backwards it doesn’t seem complicated, but looking forwards there are an enormous number of possible next actions.
Similar to finding a filament for lightbulbs, it might seem obvious to use tungsten, but at the time it wasn’t and Edison searched thousands of materials.
- N_Lens 3mo agoTesla's critique of Edison was valid though (That he would rather spend a long time empirically testing things, rather than use a bit of theory and knowledge to narrow the search field).
- jeremyjh 3mo agoYes and at 1-ply Leela will destroy Stockfish because her eval is so much better. Searching is slow even when its really fast.
- catigula 2mo agoI think the models are legitimately doing what they're good at; tireless search across an extremely large corpus of data. Humans aren't particularly good at this (in fact, they're absolutely terrible). The fact that we remain competitive and superior in many aspects isn't because we can instantly sift through tons of data, it's because we learn and correlate and have superior heuristics. In my own use, I find that AI is really good at finding bugs that are ultimately trivial but require searching through a convoluted series of inter-related files. This takes time for a human.