Yeah, I'm being very hypothetical when discussing constant money supply. For the purposes of this discussion, just assume that somehow bankers decided to not increase the money supply.
Are you in agreement then that over the long term, the total world index fund must approximate the total growth in money supply (I guess assuming constant money velocity)? If not can you help me understand why not?
Also related: can GDP increase somehow if money supply is fixed and money velocity is fixed?
Yes that makes sense, but is there some reason we should expect the total market cap to continue growing to huge multiples of the money supply (assuming continued technological improvement but fixed money supply)?
I don't understand how we can expect a basket of all publicly traded securities (something like the ticker VTWAX) to grow above the rate of global money supply growth over the long term.
Let's assume there is a fixed amount of money in the economy, and the money supply is not increasing. For the market cap of VTWAX to continue increasing, it seems like the money must be drawn away from other areas of the economy. Assuming something like 7% share price growth per year (roughly the historical average), eventually we will reach a point where the market cap of VTWAX is 100% of all money in circulation. Heck, eventually the share price... (read 152 more words →)
Very nice post. It is certainly useful to do this exercise of manually encoding language rules into the weights of a transformer in order to better understand the machinery involved.
"The ultimate ambition of this work would be to go toe-to-toe with a comparably-sized Transformer model trained in the traditional way on a modern-sized data set. This might require several people-years of focused effort though."
There is a long history of attempting to parse natural language with hand design rules and heuristics. The general consensus now is that hand engineering is insufficient, and some learning from data is necessary. To me it seems that this direction inherits the problems of these old fashioned language systems since you are codifying your own hand designed heuristics and rules into the network weights.
Do you see a way to introduce learning from data without sacrificing the interpretability that your approach provides?
Thanks for helping clear this up! That makes a lot of sense.