In non-stationary multivariate time series, different variables and samples often exhibit heterogeneous latent dynamic states, while existing deep forecasting models usually compress them into a unified end-to-end mapping, leading to suboptimal modeling of time-varying dynamics and limited interpretability regarding wh...
This work introduces Learnable Lens Networks (LLN), a physics-inspired architecture that replaces direct feature-space residual accumulation with learnable optical transport in an augmented position-angle phase space and establishes that LLN transport is globally invertible and volume-preserving for any differentiable...
Bin-Bin Yong, Zhao Su, Lan Guo et al.· 0 citations
KANFIS employs an additive aggregation mechanism, under which both model parameters and rule complexity scale linearly with input dimensionality rather than exponentially, resulting in an intrinsically interpretable model with clear rule semantics and transparent inference processes.
Bin-Bin Yong, Haoran Pei, Jun Shen et al.· arXiv.org· 3 citations
HYDRA is a parameter-efficient hyperbolic extension of KAN that combines spline-based functional learning with representations in the Poincar\'e ball, and consistently achieves competitive or superior predictive performance while improving parameter efficiency and representation interpretability.
Hyperbolic ANFIS (HyperANFIS), a hyperbolic extension of ANFIS that preserves the fuzzy semantics and core architecture of conventional ANFIS while performing rule-prototype learning, rule activation, and consequent aggregation in hyperbolic space, strengthens the fuzzy inference process.
Haoran Pei, Zhao Su, Ze-Tao Lin et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
