CAUSE: Learning Granger Causality from Event Sequences using Attribution Methods

Abstract

We study the problem of learning Granger causality between event types from asynchronous, interdependent, multi-type event sequences. Existing work suffers from either limited model flexibility or poor model explainability and thus fails to uncover Granger causality across a wide variety of event sequences with diverse event interdependency. To address these weaknesses, we propose CAUSE (Causality from AttribUtions on Sequence of Events), a novel framework for the studied task. The key idea of CAUSE is to first implicitly capture the underlying event interdependency by fitting a neural point process, and then extract from the process a Granger causality statistic using an axiomatic attribution method. Across multiple datasets riddled with diverse event interdependency, we demonstrate that CAUSE achieves superior performance on correctly inferring the inter-type Granger causality over a range of state-of-the-art methods.

References in corpus (5)

• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Axiomatic Attribution for Deep Networks
• Learning Important Features Through Propagating Activation Differences
• Tick: a Python library for statistical learning, with a particular emphasis on time-dependent modelling

Cited by in corpus (3)

• Visual Causality Analysis of Event Sequence Data
• Survey on Visual Analysis of Event Sequence Data
• An Empirical Study: Extensive Deep Temporal Point Process