Covariate Representation

Covariate representation focuses on learning informative and useful representations of input variables (covariates) for improved prediction and causal inference. Current research emphasizes disentangling covariates into factors representing confounding, instrumental, and adjustment variables, often employing techniques like variational autoencoders and generative adversarial networks within neural network architectures, including transformers adapted for time series data. These advancements aim to address challenges like confounding bias, concept drift, and the need for robust and generalizable models across diverse applications, ultimately improving the accuracy and reliability of predictions and causal effect estimations in various fields.

Papers

June 22, 2024

Learning When the Concept Shifts: Confounding, Invariance, and Dimension Reduction
Kulunu Dharmakeerthi, YoonHaeng Hur, Tengyuan Liang
LeArning Abstract Structural Causal Model Covariate Shift Alpha Invariance Dimension Reduction Relevant Covariates Confounding Bias Concept Shift Covariate Representation

June 4, 2024

Disentangled Representation via Variational AutoEncoder for Continuous Treatment Effect Estimation
Ruijing Cui, Jianbin Sun, Bingyu He, Kewei Yang, Bingfeng Ge
Variational Autoencoder Disentanglement Capability Treatment Effect Disentangled Representation Dose Response Covariate Representation

May 22, 2024

Leveraging 2D Information for Long-term Time Series Forecasting with Vanilla Transformers
Xin Cheng, Xiuying Chen, Shuqi Li, Di Luo, Xun Wang, Dongyan Zhao, Rui Yan
2 Dimensional Time Series Prediction Long Term Time Series Forecasting Vanilla Transformer Grid Based Input Time Series Covariate Representation

March 21, 2024

Contrastive Balancing Representation Learning for Heterogeneous Dose-Response Curves Estimation
Minqin Zhu, Anpeng Wu, Haoxuan Li, Ruoxuan Xiong, Bo Li, Xiaoqing Yang, Xuan Qin, Peng Zhen, Jiecheng Guo, Fei Wu, Kun Kuang
Contrastive Learning Counterfactual Prediction Dose Response Covariate Representation

October 10, 2023

iTransformer: Inverted Transformers Are Effective for Time Series Forecasting
Yong Liu, Tengge Hu, Haoran Zhang, Haixu Wu, Shiyu Wang, Lintao Ma, Mingsheng Long
Transformer Megatron Decepticons Transformer Based Transformer Architecture Time Series Forecasting Covariate Representation Linear Forecasting

July 24, 2023

De-confounding Representation Learning for Counterfactual Inference on Continuous Treatment via Generative Adversarial Network
Yonghe Zhao, Qiang Huang, Haolong Zeng, Yun Pen, Huiyan Sun
Generative Adversarial Network Counterfactual Inference Counterfactual Estimation Valued Treatment Covariate Representation Counterfactual Recurrent Network

June 23, 2023

Variance-Covariance Regularization Improves Representation Learning
Jiachen Zhu, Katrina Evtimova, Yubei Chen, Ravid Shwartz-Ziv, Yann LeCun
Transfer Learning Self Supervised Representation Learning Covariance Regularization Covariate Representation

January 27, 2023

Unearthing InSights into Mars: Unsupervised Source Separation with Limited Data
Ali Siahkoohi, Rudy Morel, Maarten V. de Hoop, Erwan Allys, Grégory Sainton, Taichi Kawamura
DCU Insight AQ Limited Data Source Separation Ill Posed Inverse Problem Signal to Signal Translation Covariate Representation

February 9, 2022

Covariate-informed Representation Learning to Prevent Posterior Collapse of iVAE
Young-geun Kim, Ying Liu, Xuexin Wei
Latent Representation Variational AutoEncoder Posterior Collapse Posterior Network Covariate Representation

January 21, 2022

Individual Treatment Effect Estimation Through Controlled Neural Network Training in Two Stages
Naveen Nair, Karthik S. Gurumoorthy, Dinesh Mandalapu
Causal Effect Estimation Individual Treatment Effect Treatment Effect Estimation Multiple Stage Covariate Representation Controllable Network Learning

January 20, 2022

DRTCI: Learning Disentangled Representations for Temporal Causal Inference
Garima Gupta, Lovekesh Vig, Gautam Shroff
Meaningful Representation Text Based Confounders Covariate Representation Counterfactual Recurrent Network

Covariate Representation

Papers

Learning When the Concept Shifts: Confounding, Invariance, and Dimension Reduction

Disentangled Representation via Variational AutoEncoder for Continuous Treatment Effect Estimation

Leveraging 2D Information for Long-term Time Series Forecasting with Vanilla Transformers

Contrastive Balancing Representation Learning for Heterogeneous Dose-Response Curves Estimation

iTransformer: Inverted Transformers Are Effective for Time Series Forecasting

De-confounding Representation Learning for Counterfactual Inference on Continuous Treatment via Generative Adversarial Network

Variance-Covariance Regularization Improves Representation Learning

Unearthing InSights into Mars: Unsupervised Source Separation with Limited Data

Covariate-informed Representation Learning to Prevent Posterior Collapse of iVAE

Individual Treatment Effect Estimation Through Controlled Neural Network Training in Two Stages

DRTCI: Learning Disentangled Representations for Temporal Causal Inference