David Arbour (@darbour26) profile picture

Causal Inference, Experiments, Networks, Machine Learning. @BHCCBoston alum.

Joined June 2009
📄 Our paper on comprehensive, low-variance evaluation of slate recommendation strategies using offline logged data was accepted at AAAI-24! Work done during my internship at @AdobeResearch with @darbour26, Georgios Theocharous, and Nikos Vlassis. (arxiv.org/abs/2308.14165)
1
6
561
David Arbour retweeted
A couple weeks ago, my, @darbour26, Tung Mai and Anup Rao’s paper on online balancing experimental design appeared in ICML! We show how to reduce variance when treatments are assigned sequentially, e.g. survey experiments or online A/B tests. icml.cc/virtual/2022/spotlig…
1
3
2
18
David Arbour retweeted
Let me be your boss! We're hiring a company expert on moderation on our brand new DS team. Reddit is growing and making big investments in our mod tools. If you're interested in helping us build the next big thing for Reddit, please dm me. boards.greenhouse.io/reddit/…
2
16
2
37
Our new working paper: tell the computer what you know (and don't know!) about a causal question w/ discrete data → automatically get most precise possible answer (bounds, or a point estimate). Joint w/ @guilhermejd1 @nsfinkelstein @dean_c_knox Shpitser.🧵arxiv.org/abs/2109.13471
7
106
42
465
Another interest talk about doubly-robust confidence sequences: drive.google.com/file/d/1bAh…
Just saw an interesting Speed talk on JSM 2021 about discretization in continuous time causal models: youtube.com/watch?v=60K3kvlR…
5
11
📢 Call for abstracts 📢 for the Conference on Digital Experimentation (CODE@MIT) Submit your best work *by Sept 15* on online field experiments, statistical methods, practical & computational challenges, etc. ide.mit.edu/events/2021-conf… #mitcode
1
16
4
29
Tonight at #ICML2021, I will be presenting my work with @purva_pruthi and David Jensen on "How and Why to Use Experimental Data to Evaluate Methods for Causal Inference". Our poster will be in spot A4 in Gather Town Room 7, tonight from 12am - 2am EDT.
1
1
13
David Arbour retweeted
🚨New!🚨 You're estimating a population mean from samples observed with varied probabilities. Do you use a Horvitz–Thompson/IPW or a Hájek/self-normalizing estimator? @Stats_samir and I examine an old question due to Trotter & Tukey (1954): why not both?arxiv.org/abs/2106.07695 🧵
5
21
3
94
David Arbour retweeted
Clever work from @iamwillcai & @jugander decomposing peer effects: When we copy others, how much is just based on observing others' choice vs. the rewards they receive from their choices? ojs.aaai.org/index.php/ICWSM…
2
5
19
David Arbour retweeted
🚨 COOL JOB ALERT 🚨: Use network sci, exploratory analysis, causal inference, surrogacy measurement @Twitter. Help improve measurement, follow-recommendations, & other problems to make twitter more friendly & better for newcomers!
2
45
1
117
David Arbour retweeted
Excited to announce that our tutorial with @darbour26 on "Causal inference from network data" has been accepted as a lecture-style tutorial at #KDD2021. Stay tuned for more details: netcause.github.io/. @kdd_news
1
18
2
127
Conformal inference gives rigorous outlier/out-of-distribution detection. We show how to control FDR with conformal p-values -- even though they are dependent, they satisfy the PRDS property! arxiv.org/abs/2104.08279 With E. Candès, @lihua_lei_stat, Y. Romano, and M. Sesia
David Arbour retweeted
Check this out 👇 our aim here is to take a closer look at the assumptions in #causalinferene! one of the most exciting aspects of this workshop is bringing in experts from epi, (bio)stats, & philosophy in a major CS conference #ML #icml2021
Excited about our @icmlconf workshop on the *Neglected Assumptions in Causal Inference* (sites.google.com/view/naci20…) with my amazing co-organizers @LauraBBalzer @alexdamour @uhlily @raziehnabi @ShalitUri We warmly welcome contributions from outside CS as well (see CFP) @ICML2021
6
17
David Arbour retweeted
These are easy to use: take your existing non/semi-parametric estimators & CIs, and replace the root-(1/n) width with these root-(log(n)/n) widths to get a time-uniform coverage guarantee for an infinite time horizon. More extensions, tweaks, etc. coming soon!
1
7
David Arbour retweeted
Unlike classical confidence intervals, confidence sequences give you the flexibility to make inferences at arbitrary (data-dependent) stopping times (e.g. in a sequential experiment or an observational study where data are collected in an online fashion).
1
3
9
David Arbour retweeted
Super excited to share this work with @edwardhkennedy, @darbour26, Aaditya Ramdas, and Ritwik Sinha on confidence sequences for causal effects! arxiv: arxiv.org/pdf/2103.06476.pdf code: github.com/wannabesmith/sequ…
4
25
1
102
David Arbour retweeted
We're continuing to work hard on methods for experimental design, so if you have any tricky problems in this arena, feel free to reach out! I'd love to help if I can. And more is coming soon :)
1
5
David Arbour retweeted
Our paper on SoftBlock (design for HTEs) will be in poster session 4 tomorrow. We have implementations in both Python and R (along with an Rmd demo using elections data). Paper: proceedings.mlr.press/v130/a… Code: github.com/ddimmery/softbloc… Previous thread: nitter.gq/DrewDim/status/1320796…
Do you run experiments? Do you care about heterogeneous treatment effects? @darbour26, Anup Rao and I have a new paper that looks specifically at how to design experiments to make sure you end up with minimal error in your HTE estimates. A thread. arxiv.org/abs/2010.11332
1
3
1
10
David Arbour retweeted
First, with My Phan and Anup Rao, we develop methods for balance in experiments when you care particularly about *generalization to a target population*. This will be in poster session 1 this afternoon at 5pm ET. Paper: proceedings.mlr.press/v130/p… Code: github.com/myphan9/Designing…
1
1
7