Fred Lu (@fredlu_1618) profile picture

Statistics, glorified statistics, and beer. MS Stanford '20

Seattle, WA
Joined January 2018
Photo by Fred Lu (@fredlu_1618)
My advisor @edwardhkennedy and I have posted an article on continuous trt effects estimation: arxiv.org/pdf/2207.11825.pdf We study nonparam models where the dose-response curve has its own smoothness, which may differ from that of the outcome regression or cond dens of trt given X.
4
17
1
93
Fred Lu retweeted
NO FUCKING WAY I JUST SEEN THIS
114
789
245
4,182
Shooting stars? Debris? Anyone else see this in Seattle just now? @KING5Seattle @komonews #UFOs
58
268
133
973
To use this video in a commercial player or in broadcasts, please email [email protected]
2
The Falcon 9 second stage from the Mar 4 Starlink launch failed to make a deorbit burn and is now reentering after 22 days in orbit. Its reentry was observed from the Seattle area at about 0400 UTC Mar 26.
268
2,058
1,241
5,984
Excited to share our #ICLR2021 paper w/ CS & math depts @Stanford 🎊 Evaluating the Disentanglement of Deep Generative Models through Manifold Topology! w/ @ericzelikman Fred Lu @AndrewYNg Gunnar Carlsson @StefanoErmon. Acknowledging @torbjornlundh Samuel Bengmark. Thread 🧵
7
64
4
380
Remember this when you buy their shitty Pokémon game for the 100th time
Nintendo will use the player brands built through our own grassroot efforts to promote their new games. Then, not only do they not help us grow, they actively prevent us from growing on our own. So tired of this company. #freemelee
42
1,305
57
10,789
Of all the shit from Nintendo this takes the prize. It's a global pandemic going on and Nintendo once again want the competitive scene to suffer Is it to much to ask that people can play and compete in games from home during this time? #FreeMelee @NintendoAmerica @Nintendo
1
Great collaboration with Gal Koplewitz, @fredlu_1618 @SKNKsmash, and Caroline Buckee now publicly available @medrxivpreprint with great insights from @majohanss
Predicting Dengue Incidence Leveraging Internet-Based Data Sources: A Case Study in 20 cities in Brazil. medrxiv.org/cgi/content/shor… #medRxiv
1
2
Shocking that we are 6 months into this #COVID__19 crisis and testing wait times are still far from helpful to meaningfully contribute to curb transmission. Great collaboration with @davidlazer @royperlis @Ognyanova @matthewabaum and j druckman
You’re unable to view this Post because this account owner limits who can view their Posts. Learn more
3
6
So important in all applied ML fields! Treat each CV fold as a simulated reality, only do preprocessing and feature selection within each simulation.
The more papers I read for a review article I'm writing about ML pitfalls in genomics, the more my faith is shaken in the results from papers that apply machine learning to methylation arrays. A salty thread. 1/
1
Important disclaimer about an article about police-involved fatalities in PNAS that made inferences using the wrong conditional probability. The "fatal"-ly incorrect analysis was published but PNAS declined to publish this letter. Sad!
New study of fatal police-involved shootings reported “no overall evidence of anti-Black ... disparities.” @dean_c_knox & I submitted a letter to the editor showing study's approach can't support claim. Journal declined to publish. Sharing here so future research can improve. 1/N
1
My open letter to @TheUSMLE & @NBMEnow re: their 6/4/20 announcement about abridged #Step1 & #Step2 CK exams only at "event-based testing" sites PLUS their claim of being "committed to ensuring equity." Like, RT to endorse. #MedTwitter #MedStudentTwitter #StepUpUSMLE #StepUpNBME
3
55
6
135
Estimating the Prevalence of COVID-19 in the United States: Three Complementary Approaches medrxiv.org/cgi/content/shor… #medRxiv
1
3
“The obligation to eliminate cannabis in countries where there was widespread traditional use is a clear example of the colonial background of the Convention, and a violation of indigenous and religious rights." narratively.com/these-enterp…
New paper alert! "Sensitivity analysis via % of unmeasured confounding" arxiv.org/abs/1912.02793 SA is absolutely crucial for causality: holy grail is finding one that's interpretable *&* flexible @bonv3 & I propose contamination model approach, w/effect bds across % confounded
4
38
3
145
Low-IQ Dems are TERRIBLE at calculus. Been trying to differentiate e^x for YEARS - keep getting SAME RESULT. #Weak leadership!! How are we going to keep up with China?? #MATH #DataScience #EconTwitter #Statistics #epitwitter #AcademicChatter #AcademicTwitter #rstats #MedTwitter