Lars van der Laan
@LarsvanderLaan3
Followers
701
Following
428
Media
32
Statuses
340
Ph.D. Student @uwstat; Research fellowship @Netflix; visiting researcher @UCJointCPH; M.A. @UCBStatistics - machine learning; semiparametrics; causal inference.
Seattle
Joined March 2022
Had a great time presenting at #ACIC on doubly robust inference via calibration. Calibrating nuisance estimates in DML protects against model misspecification and slow convergence. Just one line of code is all it takes.
2
6
68
Calibrate your outcome predictions and propensities using isotonic regression as follows:. mu_hat <- as.stepfun(isoreg(mu_hat, Y))(mu_hat) . pi_hat <- as.stepfun(isoreg(pi_hat, A))(pi_hat) . (Or use the isoreg_with_xgboost function given in the paper, which I recommend).
0
0
7
(Correction: I meant #CLeaR—Causal Learning and Reasoning.) Had a great time in Switzerland and highly recommend this conference!.
0
0
3
Thrilled to share that our work on stabilized inverse probability weighting via calibration was accepted to #CLEAR2025! I gave an oral presentation last week and was honored to receive the Best Paper Award. I’ll be giving a related poster talk at #ACIC on calibration and DML.
If you are doing causal inference with machine learning, calibrate your nuisances! More details to come….
2
0
15
Excited to share that our paper on uncertainty-aware model calibration using generalized Venn–Abers methods was accepted to #ICML2025! . We show that Vovk’s Venn–Abers calibration and conformal prediction are both special cases of a unified framework.
🚨 Excited about this new paper on Generalized Venn Calibration and conformal prediction! . We show that Venn and Venn-Abers can be extended to general losses, and that conformal prediction can be viewed as Venn multicalibration for the quantile loss! . #calibration #conformal.
3
4
28
This work is a result of my internship at Netflix over the summer and is joint with Aurelien Bibaut and Nathan Kallus. Link:
0
0
2
I’ll be giving an oral presentation at #ACIC2025 in the Advancing Causal Inference session with ML on Wednesday! . My talk is on Automatic Double Reinforcement Learning and long term causal inference. I’ll discuss Markov decision processes, Q-functions, and calibration for RL
1
1
12
🚨.
We have two papers accepted at #ICML2025! An oral on construct validity of medical LLM benchmarks and a poster on generalized Venn-Abers calibration. See you in Vancouver 🇨🇦!
0
1
8
RT @predict_addict: The paper "Generalized Venn and Venn-Abers Calibration with Applications in Conformal Prediction" by Lars van der Laan….
0
2
0
🚨 Excited about this new paper on Generalized Venn Calibration and conformal prediction! . We show that Venn and Venn-Abers can be extended to general losses, and that conformal prediction can be viewed as Venn multicalibration for the quantile loss! . #calibration #conformal.
Generalized Venn and Venn-Abers Calibration with Applications in Conformal Prediction.
0
6
20
RT @StatMLPapers: Generalized Venn and Venn-Abers Calibration with Applications in Conformal Prediction
arxiv.org
Ensuring model calibration is critical for reliable prediction, yet popular distribution-free methods such as histogram binning and isotonic regression offer only asymptotic guarantees. We...
0
3
0
Thrilled to share our new paper! We introduce a generalized autoDML framework for smooth functionals in general M-estimation problems, significantly broadening the scope of problems where automatic debiasing can be applied!. Link:
2
9
38
Excited to share our work on Double RL and long-term causal inference! This project grew out of my internship at Netflix last summer.
Automatic Double Reinforcement Learning in Semiparametric Markov Decision Processes with Applications to Long-Term Causal Inference
3
7
43
RT @StatMLPapers: Causal machine learning for heterogeneous treatment effects in the presence of missing outcome data .
arxiv.org
When estimating heterogeneous treatment effects, missing outcome data can complicate treatment effect estimation, causing certain subgroups of the population to be poorly represented. In this...
0
13
0
RT @_ahmedmalaa: In Poster Session 4 (4:30 pm - 7:30 pm PST), @LarsvanderLaan3 will present our work on combining model calibration and pre….
0
4
0
This is based on joint work with @_ahmedmalaa at @UCJointCPH . link to poster info: link to paper:.
0
1
5
Excited to present "Self-Calibrating Conformal Prediction" at #NeurIPS2024 this afternoon! Join me at the poster session to learn how combining model calibration with predictive inference gives calibrated point predictions and conditionally valid prediction intervals. #conformal
4
12
99
Follow me on bluesky with the same handle Larsvanderlaan3!.
0
0
1