AI RESEARCH
Autocorrelation Reintroduces Spectral Bias in KANs for Time Series Forecasting
arXiv CS.LG
•
ArXi:2604.23518v1 Announce Type: new Existing theory suggests that Kolmogoro-Arnold Networks (KANs) can overcome the spectral bias commonly observed in neural networks under the assumption that inputs are statistically independent. However, this assumption does not hold in time series forecasting (TSF), where inputs are lagged observations with strong temporal autocorrelation. Through theoretical analysis and empirical validation, we obtain an unexpected finding: temporal autocorrelation re