> It’s also a rare time in history that an individual can make lots of progress. Most of us need to be a part of big groups to do anything worthwhile, in most fields. But in this case lone wolves often have the upper hand over established organizations.
I'm curious why you think this is true? My feeling as a broke individual trying to catch up on ml is that there are some simple demos to do. But scaling up requires a lot of compute and storage for an individual. Acquiring datasets and training are cost prohibitive. I'm only able to play around with some really small stuff because by dumb luck a few years ago I bought a gaming laptop with a nvidia gpu in it. The impressive models that are generating the hype are just a different league. Love to hear to how I am wrong though?
It’s true that you need compute to do large experiments, but the large experiments grow out of small ones. If you can show promising work in a small way, it’s easier to get compute. You can also apply to TRC to get a bunch of TPUs. They had capacity issues for a long time but I’ve heard it’s improved.
Don’t focus on the hype models. Find a niche that you personally like, and do that. If you’re chasing hype you’ll always be skating towards the puck. My original AI interest was to use voice generation to make Dr Kleiner sing about being a modern major general. It went from there, to image gen, to text gen, and kaboom, the whole world blew up. I was the first to show that GPTs can be used for more than just language modeling — in my case, playing chess.
Wacky ideas like that are important to play around with, because they won’t seem so wacky in a year.
Thats interesting, at a glance the TRC thing looks more altruistic and impactful than what I had in mind for learning or making money. I'll have to keep it in mind if I do, do something share worthy ever. Thanks!
In a gold rush, sell shovels! The ML pipeline has a lot of bottlenecks. Work on one, get useful and novel expertise, and have a massive impact on the industry. Like maybe you could find a way to optimise your GPU usage? Is there a way to package what you feed it more efficiently?
The point not being of competing with OpenAI, but to solve a problem that everyone in the field has.
Another area where there's potential for an individual to make lots of progress is in theory and mechanistic interpretation. Although it's not where the money is, it's probably not rapid progress and it's really hard.
I'm curious why you think this is true? My feeling as a broke individual trying to catch up on ml is that there are some simple demos to do. But scaling up requires a lot of compute and storage for an individual. Acquiring datasets and training are cost prohibitive. I'm only able to play around with some really small stuff because by dumb luck a few years ago I bought a gaming laptop with a nvidia gpu in it. The impressive models that are generating the hype are just a different league. Love to hear to how I am wrong though?