The paper proposes contrastive trend estimation (CTE) as a universal method for identifying hidden monotonic trends in any type of temporal data without standard assumptions.
Unsupervised feature extraction by time- contrastive learning and nonlinear ica
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2022 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Universal hidden monotonic trend estimation with contrastive learning
The paper proposes contrastive trend estimation (CTE) as a universal method for identifying hidden monotonic trends in any type of temporal data without standard assumptions.