NeRT factorizes periodic signals into a sine-based periodic factor and an unbounded scale factor, enabling extrapolation beyond the training range on several periodic benchmarks.
Therefore, the window size is a critical hyperparameter in modeling time series
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Neural Functions for Learning Periodic Signal
NeRT factorizes periodic signals into a sine-based periodic factor and an unbounded scale factor, enabling extrapolation beyond the training range on several periodic benchmarks.