Citation:
Saghafian, S. Working Paper. “Ambiguous Dynamic Treatment Regimes: A Reinforcement Learning Approach”. Copy at https://tinyurl.com/y955hbvo
 | 848 KB |
Recent Publications
- Ambiguous Dynamic Treatment Regimes: A Reinforcement Learning Approach
- The Impact of Vertical Integration on Physician Behavior and Healthcare Delivery: Evidence from Gastroenterology Practices
- COVID-19: Health and Economic Impacts of Societal Intervention Policies in the U.S.
- Who Should See the Patient? On Deviations from Preferred Patient-Provider Assignments in Hospitals
- Joint Patient Selection and Scheduling under No-Shows: Theory and Application in Proton Therapy
- Do Independent and Parallel Processing Physicians Influence Each Other's Performance? Evidence from the Emergency Department
Soroush Saghafian on Twitter
-
Enjoyed executive teaching at the "Program for Chairs of Clinical Services" @HarvardChanSPH @HarvardChanECPE . Great group of motivated individuals, and a fantastic executive teaching program! t.co/yBy5LtmalT CC: @decisionscience t.co/hhTAjTiRbg
-
@elonmusk Just fyi this a 1935 book. You might find it useful. t.co/1C8L8x0WDc
-
Random sample of 100 users to make decisions a/b a $43 billion bid that could instead be used to save so many lives, remove a good portion of humans out of poverty, and much more? What a sad day for #AnalyticsScience and its potential #public #impact. t.co/Kax9U08bt9
-
175% error! This is the error you might have if you purely rely on #COVID19 #data. Exactly why we need #models not just data! Thank you @WHO for highlighting this! t.co/7FxeGVzqvw
-
Thank you organizers of "Frontiers of Big Data, AI and Analytics" for inviting me! YouTube video of my talk: t.co/V31jdhlzws t.co/SrOA4Kuv0k
-
👇Twitter account of our research lab. Follow if you are interested.👇 @HKS_BizGov @harvard_data @decisionscience @SusanMurphylab1 @HKS_MLD_Area @INFORMS t.co/aYRjqbkyRu
-
Free webinar. Join if you are interested. #AI #MachineLearning #PublicPolicy #BigData #DecsionMaking t.co/CAz6mDkOed
-
@ArshyaFeizi @Harvard @AgniOrfanoudaki You are a fantastic addition to my lab!
-
@ArshyaFeizi @Harvard @AgniOrfanoudaki Welcome to the #PIASLab @ArshyaFeizi. I’m thrilled to have you in the lab.
-
Thank you @StanfordGSB OIT for hosting me this week as a speaker. I enjoyed my visit! Back to work now. t.co/Lh5c8VUsPG
-
How do #hospitalclosures in the United States impact patient care? See the Q&A below. t.co/Ez4afe9bf1
-
👇👇👇See the press release below if you care about #hospitals and want to know what our data shows👇👇👇 t.co/ynUqMl46vi
-
@EMDocBaker @DrDaronAcemoglu Agreed! In the paper below, we've studied how #MD vs #DO affects performance using data from the Mayo Clinic. But there is room for more comprehensive research. "Who Is an Effective and Efficient Physician? Evidence from Emergency Medicine" t.co/ksTnIjKJwE
-
Convincing paper by @DrDaronAcemoglu et al. showing managers with a #Bschool degree (e.g., #MBA) perform worse than others. They also do not share wages with their workers. Great evidence of the need to reform & improve Bschool #education? t.co/XEnfOxPG8p t.co/rUfBT8pV7C
-
👇👇👇 Our lates piece w/ @AliRaja_MD of @MassGeneralEM & my former PhD student @lina_songs. What should #policymakers and researchers know & do about #HospitalClosures? Answers here👇👇👇 #PIASLab @decisionscience @HKS_BizGov @Kennedy_School t.co/dRzeKte3QO
-
👇👇👇 Tomorrow @ 12:10pm. Join if you are interested 👇👇👇 t.co/bDrcmDqZRo
-
Our latest article "Employee Health Plans Powered by Analytics" #Analytics #MachineLearning #Healthcare t.co/hP3TLAzyUG via @MBR Journal
-
My latest talk with comments from one the most insightful and influential scholars of our time, Prof. Richard Zeckhauser. t.co/6gXN8xb4Mh t.co/7mnzm65pKC
-
My incredibly thoughtful and prolific colleague, Richard Zeckhauser (t.co/luqXRZQ2TG), will be giving opening comments and serve as the discussant.
-
Free webinar. Join if you are interested. t.co/5aGqzcoYDp
-
Announcement: New Postdoctoral Fellow Position in my lab at Harvard University. Apply if you are eligible. More info: t.co/4B7rz6VURT PIAS-Lab: t.co/blqqLMJxmh #PIASLab #postdoc #ORMS #Healthcare #Analytics t.co/b4y6GWaoPc
-
I enjoyed my interview with @INFORMS @ORMS_Tomorrow. Great discussions a/b role of #orms & #MachineLearning in #Healthcare #Analytics & #Policy, & their impact on Diversity, Equity, & Inclusion (#DEI). @decisionscience @harvard_data @HarvardBiz t.co/0x43EOCb3B
-
Happy New Year from the #PIASLAB @Harvard and some updates! See: t.co/Y9Mbvaouq5 Wishing everyone a great 2022!
-
@Harvard decided to shut down the campus (again), and campus childcare. Working without childcare support is not working. t.co/xU1Dg9PyEu
-
P.S. For interested readers, some related non-technical descriptions can be found in my blog post: t.co/d2fx36O6Pk
-
Interested in the intersection of #ReinforcementLearning, sequential #Decsion-Making & #CausalInference? This paper at the #PIAS-Lab introduces Ambiguous Dynamic Treatment Regimes (ADTRs). Causal inference should neither be model-based nor model-free. t.co/Ep9rGjtZq2
