Go to AAAI 2025 Workshop AICT homepageMCNS: Mining Causal Natural Structures Inside Time Series via A Novel Internal Causality Scheme
Keywords: Causal
Abstract: Causal inference permits us to discover covert relationships
of various variables in time series. However, in most exist-
ing works, the variables mentioned above are the dimensions.
The causality between dimensions could be cursory, which
hinders the comprehension of the internal relationship and the
benefit of the causal graph to the neural networks (NNs). In
this paper, we find that causality exists not only outside but
also inside the time series because it implies the succession
of events in the real world. It inspires us to seek the relation-
ship between internal subsequences. However, the challenges
are the hardship of discovering causality from subsequences
and utilizing the causal natural structures to improve Neural
Networks. To address these challenges, we propose a novel
framework called Mining Causal Natural Structure (MCNS ),
which is automatic and domain-agnostic and helps to find
the causal natural structures inside time series via the inter-
nal causality scheme. We evaluate the MCNS framework and
integrate NN with MCNS on time series classification tasks.
Experimental results illustrate that our impregnation, by re-
fining attention, shape selection classification, and pruning
datasets, drives NN, even the data itself preferable accuracy
and interpretability. Besides, MCNS provides an in-depth,
solid summary of the time series and datasets.
Submission Number: 7
Loading