Samir Khan

Stats PhD student at Stanford, formerly at UChicago, interested in causal inference and social science applications

[John Cherian @jjcherian]

Large language model validity via enhanced conformal prediction methods

Excited to share my work with Isaac Gibbs (@iswgibbs) and Emmanuel Candès https://t.co/muodAiM1Wj.

Question: how can we provide practical and rigorous validity guarantees for LLMs? (1/12)

[Samir Khan @stats_samir]

See the arXiv draft for more theoretical results, simulations showing how our bounds can be used to inform experimental designs, and comparison of complete randomization vs. Bernoulli randomization!
https://t.co/WcyjpLHoIK

[Samir Khan @stats_samir]

Excited to share new work with @jugander, Matt Eichhorn, and Christina Lee Yu on combining experimental design and outcome modeling for causal inference under network interference

https://t.co/WcyjpLHoIK

[Samir Khan @stats_samir]

More broadly, these results suggest that rather than asking “what is a good estimator for a Bernoulli design” or “what is a good design for an IPW estimator” we should be asking “what are good estimator/design pairs”

[Martin Saveski @msaveski]

[Please RT]
I’m recruiting PhD students to work with me at @UW!

I’m looking for students passionate about using computational methods to study how social platforms can be reimagined to enable better conversations, bridge political divides, and reduce the spread of misinfo.

>>>

[Davide Viviano @VivianoDavide]

New working paper with a fantastic team of coauthors:
we study when and how to design cluster experiments with network spillovers.

https://t.co/1l676HeCGV

with @lihua_lei_stat , @guido_imbens, Brian Karrer, Okke Schrijvers, Liang Shi

[Samir Khan @stats_samir]

@kylefbutts You can get something similar to this using theme_economist from the ggthemes package

https://t.co/jYzHek0oHV

[Samir Khan @stats_samir]

@Apoorva__Lal yes! everything reinvented everywhere all the time

[Samir Khan @stats_samir]

A thread about CUPAC/CUPED - the tl;dr is that "CUPAC" is nearly equivalent to G-computation/Oaxaca-Blinder/imputation methods with a somewhat unusual choice of imputation model

https://t.co/tKF3JlfUdo

1/10

[Samir Khan @stats_samir]

@VincentAB very neat, thanks!

[Samir Khan @stats_samir]

So CUPAC is intimately connected to certain other methods for ATE estimation in experiments, and these connections also suggest possible extensions of/improvements to CUPAC drawing on the literature around those other methods,

10/10

[Samir Khan @stats_samir]

Similarly, it models Y_i(1) by taking the non-parametric model for Y_i(0), and learning Y_i(1) as a linear transformation of Y_i(0) on the experimental data - note that this relies on Y_i(1) being an approximately linear function of Y_i(0)

9/10