Mining Causality from Continuous-time Dynamics Models: An Application to Tsunami Forecasting
This addresses the need for interpretable causal insights in domains like tsunami forecasting, where understanding relationships is as crucial as predictions, though it is incremental as it builds on existing continuous-time models.
The paper tackles the problem of identifying causal structures from continuous-time dynamics models, which are opaque due to neural network parameterization, and applies it to tsunami forecasting. The proposed method enforces sparsity in input layer weights to learn physically-consistent causal relationships while maintaining high forecasting accuracy.
Continuous-time dynamics models, such as neural ordinary differential equations, have enabled the modeling of underlying dynamics in time-series data and accurate forecasting. However, parameterization of dynamics using a neural network makes it difficult for humans to identify causal structures in the data. In consequence, this opaqueness hinders the use of these models in the domains where capturing causal relationships carries the same importance as accurate predictions, e.g., tsunami forecasting. In this paper, we address this challenge by proposing a mechanism for mining causal structures from continuous-time models. We train models to capture the causal structure by enforcing sparsity in the weights of the input layers of the dynamics models. We first verify the effectiveness of our method in the scenario where the exact causal-structures of time-series are known as a priori. We next apply our method to a real-world problem, namely tsunami forecasting, where the exact causal-structures are difficult to characterize. Experimental results show that the proposed method is effective in learning physically-consistent causal relationships while achieving high forecasting accuracy.