A tiny KAN forecasts the VIX as accurately as large MLPs, but its symbolified output is a linear autoregression that mirrors the classic HAR model.
Cybenko, Approximation by superpositions of a sigmoidal function, Mathematics of Control, Signals and Systems 2 (1989) 303–314
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Forecasting VIX using interpretable Kolmogorov-Arnold networks
A tiny KAN forecasts the VIX as accurately as large MLPs, but its symbolified output is a linear autoregression that mirrors the classic HAR model.