Research map / Learning theory and algorithms / Kernels and regression theory

Statistical Mixing Guarantees for Contractive Echo State Networks

NeurIPS 2026 · Learning theory and algorithms › Kernels and regression theory · approach: High-dimensional asymptotic generalization

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Develops a transport-based learning theory for contractive echo state networks with stochastic inputs, treating the reservoir as a Markov process and proving Wasserstein contraction to give statistical mixing guarantees for learning from a single dependent trajectory.

Problem

Reservoir computing features come from a single highly dependent trajectory, so the effective amount of usable data for learning readout regressions is unclear.

Approach

Views a contractive echo state network driven by stochastic inputs as a Markov process and proves a Wasserstein contraction, which yields statistical guarantees for learning from dependent reservoir features via mixing.

Closest papers by meaning

• Frequency Domain Reservoir Computing (NeurIPS 2026 · similarity 0.89)
• Universality of Real Minimal Complexity Reservoir (AAAI 2025 · similarity 0.86)
• Spectral-Aware Reservoir Computing for Fast and Accurate Time Series Classification (ICML 2025 · similarity 0.85)
• Sequence Modeling with Spectral Mean Flows (NeurIPS 2025 · similarity 0.85)
• Learning long range dependencies through time reversal symmetry breaking (NeurIPS 2025 · similarity 0.84)
• Deep Learning for Computing Convergence Rates of Markov Chains (NeurIPS 2024 · similarity 0.84)