Lightnews — Scholar-powered news

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 7h

"investigative applied theory" would be a cool field

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 7h

every feed is another lever you can pull. sometimes the lever has hooks to keep you pulling. the only way you find out is by pulling. at least the hooks here are probably unintentional, for now.

1 2

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 7h

a recsys that maximizes time on app is kind of a hostile thing to build when you think about it. it also seems like the kind of thing that would emerge from extended use + algorithmic diversity + directed search

1 9

Reposted by grand theft eigenvalue 🔆

ponder @ponder.ooo · 8h

slime mold

projected commodity flow by trucks in the usa in 2050

4 1 14

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 8h

i keep hoping one of them will figure that out! alas / fortunately

2

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 9h

🙏🙏🙏

1

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 9h

senior cat can't decide if she's concerned or mad

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 9h

it's extremely satisfying to watch the housecleaning robot learn where things are

1 13

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 9h

no AI or capitalism here. just good old fashioned computer vision, robotics, and trade

1 6

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 9h

it's extremely satisfying to watch the housecleaning robot learn where things are

1 1 13

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 9h

like this is in the direction i'm after, but i swear i've seen something more recent that compared logit/mlogit/etc against like random forest and some nn things

Demand Modeling, Forecasting, and Counterfactuals, Part I

Founded in 1920, the NBER is a private, non-profit, non-partisan organization dedicated to conducting economic research and to disseminating research findings among academics, public policy makers, an...

www.nber.org

1 4

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 9h

any econometricians seeing this know of recent-ish papers comparing out-of-sample predictive performance of discrete choice models? particularly interested in anything getting at choice data that can't be rationalized with the variables available

1 3 9

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 11h

great ai boyfriend panic of 2027

1

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 11h

"a straussian read will tell you that you're probably mood affiliating and not updating your priors" i said as i slowly turned into a mushroom

1 6

Reposted by grand theft eigenvalue 🔆

Craig @craigplazure.bsky.social · 9d

Tyler Cowen the food blogger?

1 4

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 15h

Be Loud. Create Large Residuals.

3 9

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 15h

"be rational" is how They tell You to Take Up Less Space

1 1 3

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 15h

economic rationality is a compression scheme

1 1

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 15h

tired: macroeconomist

wired: aggregation sociologist

1 6

grand theft eigenvalue 🔆 @akhilrao.bsky.social · 16h

"is the archipelago of gcs and discords in which i repeatedly interact with ~20 people an echo chamber" greatest thread in the history of manic gc posts

4