Conservative Policy Construction Using Variational Autoencoders for Logged Data with Missing Values

09/08/2021
by   Mahed Abroshan, et al.
11

In high-stakes applications of data-driven decision making like healthcare, it is of paramount importance to learn a policy that maximizes the reward while avoiding potentially dangerous actions when there is uncertainty. There are two main challenges usually associated with this problem. Firstly, learning through online exploration is not possible due to the critical nature of such applications. Therefore, we need to resort to observational datasets with no counterfactuals. Secondly, such datasets are usually imperfect, additionally cursed with missing values in the attributes of features. In this paper, we consider the problem of constructing personalized policies using logged data when there are missing values in the attributes of features in both training and test data. The goal is to recommend an action (treatment) when , a degraded version of with missing values, is observed. We consider three strategies for dealing with missingness. In particular, we introduce the conservative strategy where the policy is designed to safely handle the uncertainty due to missingness. In order to implement this strategy we need to estimate posterior distribution p(|), we use variational autoencoder to achieve this. In particular, our method is based on partial variational autoencoders (PVAE) which are designed to capture the underlying structure of features with missing values.

READ FULL TEXT
research
09/17/2021

Understanding the Effects of Visualizing Missing Values on Visual Data Exploration

When performing data analysis, people often confront data sets containin...
research
08/13/2016

An approach to dealing with missing values in heterogeneous data using k-nearest neighbors

Techniques such as clusterization, neural networks and decision making u...
research
02/19/2019

On the consistency of supervised learning with missing values

In many application settings, the data are plagued with missing features...
research
04/21/2022

Interpolation of Missing Swaption Volatility Data using Gibbs Sampling on Variational Autoencoders

Albeit of crucial interest for both financial practitioners and research...
research
03/02/2022

Learning Conditional Variational Autoencoders with Missing Covariates

Conditional variational autoencoders (CVAEs) are versatile deep generati...
research
06/30/2023

Thompson sampling for improved exploration in GFlowNets

Generative flow networks (GFlowNets) are amortized variational inference...
research
12/23/2016

Constructing Effective Personalized Policies Using Counterfactual Inference from Biased Data Sets with Many Features

This paper proposes a novel approach for constructing effective personal...

Please sign up or login with your details

Forgot password? Click here to reset