Paper Image

Inferring causal relationships from time series data

Published on:

10 October 2023

Primary Category:

Machine Learning

Paper Authors:

Sumanth Varambally,

Yi-An Ma,

Rose Yu

Bullets

Key Details

Proposes method to learn causal graphs from time series data

Can handle data from mixtures of causal models

Infers cluster membership and multiple causal graphs

Accounts for nonlinearity and history-dependent noise

Shows improved performance over other methods

AI generated summary

Inferring causal relationships from time series data

This paper proposes a method to infer causal relationships from time series data, even when the data comes from a mixture of different underlying causal models. The method can learn multiple causal graphs corresponding to different data clusters, determine which samples belong to which cluster, and handle complex relationships like nonlinearity and history-dependent noise.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up