Exploring Neural Granger Causality with xLSTMs: Unveiling Temporal Dependencies in Complex Data
This work addresses the problem of identifying temporal dependencies in complex data for researchers in time series analysis, but it is incremental as it builds on existing xLSTM architectures.
The paper tackled the challenge of capturing long-range dependencies in Granger causality analysis for time series by proposing GC-xLSTM, which uses a novel dynamic loss penalty and joint optimization to enforce sparsity and robustly recover causal relations, achieving improved performance on six diverse datasets.
Causality in time series can be challenging to determine, especially in the presence of non-linear dependencies. Granger causality helps analyze potential relationships between variables, thereby offering a method to determine whether one time series can predict-Granger cause-future values of another. Although successful, Granger causal methods still struggle with capturing long-range relations between variables. To this end, we leverage the recently successful Extended Long Short-Term Memory (xLSTM) architecture and propose Granger causal xLSTMs (GC-xLSTM). It first enforces sparsity between the time series components by using a novel dynamic loss penalty on the initial projection. Specifically, we adaptively improve the model and identify sparsity candidates. Our joint optimization procedure then ensures that the Granger causal relations are recovered robustly. Our experimental evaluation on six diverse datasets demonstrates the overall efficacy of GC-xLSTM.