Recursive Partitioning for Personalization using Observational Data

被引:0
|
作者
Kallus, Nathan [1 ,2 ]
机构
[1] Cornell Univ, Sch Operat Res & Informat Engn, Ithaca, NY 14853 USA
[2] Cornell Univ, Cornell Tech, Ithaca, NY 14853 USA
基金
美国国家科学基金会;
关键词
PROPENSITY SCORE; POLYMORPHISMS;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We study the problem of learning to choose from m discrete treatment options (e.g., news item or medical drug) the one with best causal effect for a particular instance (e.g., user or patient) where the training data consists of passive observations of covariates, treatment, and the outcome of the treatment. The standard approach to this problem is regress and compare: split the training data by treatment, fit a regression model in each split, and, for a new instance, predict all m outcomes and pick the best. By reformulating the problem as a single learning task rather than m separate ones, we propose a new approach based on recursively partitioning the data into regimes where different treatments are optimal. We extend this approach to an optimal partitioning approach that finds a globally optimal partition, achieving a compact, interpretable, and impactful personalization model. We develop new tools for validating and evaluating personalization models on observational data and use these to demonstrate the power of our novel approaches in a personalized medicine and a job training application.
引用
收藏
页数:10
相关论文
共 50 条
  • [1] Using recursive partitioning to analyze a large sar data set
    Young, SS
    Hawkins, DM
    [J]. SAR AND QSAR IN ENVIRONMENTAL RESEARCH, 1998, 8 (3-4) : 183 - 193
  • [2] Predicting bankruptcy using recursive partitioning and a realistically proportioned data set
    McKee, TE
    Greenstein, M
    [J]. JOURNAL OF FORECASTING, 2000, 19 (03) : 219 - 230
  • [3] Recursive partitioning on incomplete data using surrogate decisions and multiple imputation
    Hapfelmeier, A.
    Hothorn, T.
    Ulm, K.
    [J]. COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2012, 56 (06) : 1552 - 1565
  • [4] Data access skipping for recursive partitioning methods
    Kislal, Orhan
    Kandemir, Mahmut T.
    [J]. COMPUTER LANGUAGES SYSTEMS & STRUCTURES, 2018, 53 : 143 - 162
  • [5] Canonical variates for recursive partitioning in data mining
    Cappelli, C
    Conversano, C
    [J]. COMPSTAT 2002: PROCEEDINGS IN COMPUTATIONAL STATISTICS, 2002, : 213 - 218
  • [6] Recursive partitioning models for linkage in COGA data
    Xu, W
    Taylor, C
    Veenstra, J
    Bull, SB
    Corey, M
    Greenwood, CMT
    [J]. BMC GENETICS, 2005, 6 (Suppl 1)
  • [7] Recursive partitioning models for linkage in COGA data
    Wei Xu
    Chelsea Taylor
    Justin Veenstra
    Shelley B Bull
    Mary Corey
    Celia MT Greenwood
    [J]. BMC Genetics, 6
  • [8] Identifying clusters in genomics data by recursive partitioning
    Nilsen, Gro
    Borgan, Ornulf
    Liestol, Knut
    Lingjaerde, Ole Christian
    [J]. STATISTICAL APPLICATIONS IN GENETICS AND MOLECULAR BIOLOGY, 2013, 12 (05) : 637 - 652
  • [9] Analysis of a large, high-throughput screening data using recursive partitioning
    Young, SS
    Sacks, J
    [J]. MOLECULAR MODELING AND PREDICTION OF BIOACTIVITY, 2000, : 149 - 156
  • [10] Analysis of a large structure-activity data set using recursive partitioning
    Hawkins, DM
    Young, SS
    Rusinko, A
    [J]. QUANTITATIVE STRUCTURE-ACTIVITY RELATIONSHIPS, 1997, 16 (04): : 296 - 302