AB实验的高端玩法系列1 - AB实验人群定向/个体效果差异/HTE/Uplift Model 论文github收藏

一直以来机器学习希望解决的一个问题就是'what if',也就是决策指导:

  • 如果我给用户发优惠券用户会留下来么?
  • 如果患者服了这个药血压会降低么?
  • 如果APP增加这个功能会增加用户的使用时长么?
  • 如果实施这个货币政策对有效提振经济么?

这类问题之所以难以解决是因为ground truth在现实中是观测不到的,一个已经服了药的患者血压降低但我们无从知道在同一时刻如果他没有服药血压是不是也会降低。

这个时候做分析的同学应该会说我们做AB实验!我们估计整体差异,显著就是有效,不显著就是无效。但我们能做的只有这些么?

当然不是!因为每个个体都是不同的!整体无效不意味着局部群体无效!

  • 如果只有5%的用户对发优惠券敏感,我们能只触达这些用户么?或者不同用户对优惠券敏感的阈值不同,如何通过调整优惠券的阈值吸引更多的用户?
  • 如果降压药只对有特殊症状的患者有效,我们该如何找到这些患者?
  • APP的新功能部分用户不喜欢,部分用户很喜欢,我能通过比较这些用户的差异找到改进这个新功能的方向么?

以下方法从不同的角度尝试解决这个问题,但基本思路是一致的:我们无法观测到每个用户的treatment effect,但我们可以找到一群相似用户来估计实验对他们的影响。

我会在之后的博客中,从CasualTree的第二篇Recursive partitioning for heterogeneous causal effects开始梳理下述方法中的异同。

整个领域还在发展中,几个开源代码都刚release不久,所以这个博客也会持续更新。如果大家看到好的文章和工程实现也欢迎在下面评论~

Uplift Modelling/Causal Tree

  1. Nicholas J Radcliffe and Patrick D Surry. Real-world uplift modelling with significance based uplift trees. White Paper TR-2011-1, Stochastic Solutions, 2011.[文章链接]
  2. Rzepakowski, P. and Jaroszewicz, S., 2012. Decision trees for uplift modeling with single and multiple treatments. Knowledge and Information Systems, 32(2), pp.303-327.[文章链接]
  3. Yan Zhao, Xiao Fang, and David Simchi-Levi. Uplift modeling with multiple treatments and general response types. Proceedings of the 2017 SIAM International Conference on Data Mining, SIAM, 2017. [文章链接] [Github链接]
  4. Athey, S., and Imbens, G. W. 2015. Machine learning methods for
    estimating heterogeneous causal effects. stat 1050(5) [文章链接]
  5. Athey, S., and Imbens, G. 2016. Recursive partitioning for heterogeneous causal effects. Proceedings of the National Academy of
    Sciences. [文章链接] [Github链接]
  6. C. Tran and E. Zheleva, “Learning triggers for heterogeneous treatment effects,” in Proceedings of the AAAI Conference on Artificial Intelligence, 2019 [文章链接] [Github链接]

Forest Based Estimators

  1. Wager, S. & Athey, S. (2018). Estimation and inference of heterogeneous treatment effects using random forests. Journal of the American Statistical Association .
  2. M. Oprescu, V. Syrgkanis and Z. S. Wu. Orthogonal Random Forest for Causal Inference. Proceedings of the 36th International Conference on Machine Learning (ICML), 2019 [文章链接] [GitHub链接]

Double Machine Learning

  1. V. Chernozhukov, D. Chetverikov, M. Demirer, E. Duflo, C. Hansen, and a. W. Newey. Double Machine Learning for Treatment and Causal Parameters. ArXiv e-prints [文章链接] [Github链接]
  2. V. Chernozhukov, M. Goldman, V. Semenova, and M. Taddy. Orthogonal Machine Learning for Demand Estimation: High Dimensional Causal Inference in Dynamic Panels. ArXiv e-prints, December 2017.
  3. V. Chernozhukov, D. Nekipelov, V. Semenova, and V. Syrgkanis. Two-Stage Estimation with a High-Dimensional Second Stage. 2018.
  4. X. Nie and S. Wager. Quasi-Oracle Estimation of Heterogeneous Treatment Effects. arXiv preprint arXiv:1712.04912, 2017.[文章连接]
  5. D. Foster and V. Syrgkanis. Orthogonal Statistical Learning. arXiv preprint arXiv:1901.09036, 2019 [文章链接]

Meta Learner

  1. C. Manahan, 2005. A proportional hazards approach to campaign list selection. In SAS User Group International (SUGI) 30 Proceedings.
  2. Green DP, Kern HL (2012) Modeling heteroge-neous treatment effects in survey experiments with Bayesian additive regression trees. Public OpinionQuarterly 76(3):491–511.
  3. Sören R. Künzel, Jasjeet S. Sekhon, Peter J. Bickel, and Bin Yu. Metalearners for estimating heterogeneous treatment effects using machine learning. Proceedings of the National Academy of Sciences, 2019. [文章链接] [GitHub链接]

Deep Learning

  1. Fredrik D. Johansson, U. Shalit, D. Sontag.ICML (2016). Learning Representations for Counterfactual Inference [文章链接]
  2. Shalit, U., Johansson, F. D., & Sontag, D. ICML (2017). Estimating individual treatment effect: generalization bounds and algorithms. Proceedings of the 34th International Conference on Machine Learning [文章链接]
  3. Christos Louizos, U. Shalit, J. Mooij, D. Sontag, R. Zemel, M. Welling.NIPS (2017). Causal Effect Inference with Deep Latent-Variable Models [文章链接]
  4. Alaa, A. M., Weisz, M., & van der Schaar, M. (2017). Deep Counterfactual Networks with Propensity-Dropout [文章链接]
  5. Shi, C., Blei, D. M., & Veitch, V. NeurIPS (2019). Adapting Neural Networks for the Estimation of Treatment Effects
    [文章链接] [Github链接]

Uber专场

最早就是uber的博客在茫茫paper的海洋中帮我找到了方向,如今听说它们AI LAB要解散了有些伤感,作为HTE最多star的开源方,它们值得拥有一个part

  1. Shuyang Du, James Lee, Farzin Ghaffarizadeh, 2017, Improve User Retention with Causal Learning [文章连接]
  2. Zhenyu Zhao, Totte Harinen, 2020, Uplift Modeling for Multiple Treatments with Cost [文章连接]
  3. Will Y. Zou, Smitha Shyam, Michael Mui, Mingshi Wang, 2020, Learning Continuous Treatment Policy and Bipartite Embeddings for Matching with Heterogeneous Causal Effects
    Optimization [文章链接]
  4. Will Y. Zou,Shuyang Du,James Lee,Jan Pedersen, 2020, Heterogeneous Causal Learning for Effectiveness Optimization
    in User Marketing [文章连接]

想看更多因果推理AB实验相关paper的小伙伴看过来 Paper_CausalInference_abtest

持续更新中 ~

posted @ 2019-10-21 07:01  风雨中的小七  阅读(13038)  评论(2编辑  收藏  举报