Post #3080261

2026-05-25 10:39 UTC

@mc@mathstodon.xyz This also plays into my conspiracy theory that quant finance was a decade ahead of the rest of the world in deep RL when it peaked around 2017 (aka they knew some stuff back then that nobody knows to this day), but then were completely blindsided by the transformer architecture (which for them is probably far more useful to run on raw time series data than on language)

Replies (2)

@SvenGeier@mathstodon.xyz 2026-05-26 02:46

@julesh@mathstodon.xyz I keep waiting for this to really hit: if llms are so good at finding subtle patterns in data, then they should be able to make 99% of quants jobless, which would mean real estate in New York and London would suddenly get cheap. ( Until recently I'd thought retiring to London would be nice; these days I'm not so sure any more...) @mc@mathstodon.xyz

Open ##3080262
@fogti@chaos.social 2026-05-26 10:07

@julesh@mathstodon.xyz @mc@mathstodon.xyz Yeah, the application of throwing it on time series data are quite obvious. Although stocks data is problematic due to having the unit root property ( https://en.wikipedia.org/wiki/Unit_root ) after one has thrown some normalization on the data to get rid of the inflation trend and recurring trends like quartal periodic stuff (e.g. earning reports).

Open ##3080265