- Rating
- 1
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
CAW-KAN is a dual grid wavelet + KAN based model for time series forecasting. This is a channel independent forecaster which first applies revin and conv 1D layers for embedding, then dense temporal Conv 1D + Wavelet-KAN blocks sequentially, finally a projector to obtain final predictions. KAN uses a fixed dual-grid dictionary in latent value space with CP-factored coefficients. Authors reported best/tied results in multiple datasets and baselines.
Critical strengths
- Paper includes experiments on multiple forecast lengths, seeds and datasets which increases the breadth of evaluation and statistical significance.
- A very simple way of explaining the architecture and experiments are provided with matched loss results.
Critical weaknesses
- Very limited architectural novelty: The normalisation (RevIN), Conv1D based mixing, channel independent processing all the methods are from well known previous papers like TimeKAN etc. Authors just added a Wavelet-KAN blocks which is very incremental
- Limited performance gains than baselines: Authors have shown handpicked and single seed results to show in main paper, Table 10 clearly says performance gains are very less compared to baselines like HaKAN and TimeKAN. In addition even in table 1, proposed method is better only in few settings.
Questions for authors
- Why Without Wavelet-KAN block ablation is not available in paper ?
- Authors mentioned, CAW-KAN trains with Log-Cosh and the baselines with MSE is a fair comparison, how is that a fair comparison ?
- Why a handpicked and single seed results are only shown in main paper ?
- in ablation KAN to MLP (Table 2), is that a parameter matched MLP ?