Finite-Horizon Fisher Memory in Two-Sided Power-Bounded Recurrent Systems

cs.LG

Submitted: 2026-09-30

Updated: 2026-09-30

Comments: 34 pages, 7 figures. Reproducibility materials: https://github.com/jeonghoon-ad/finite-horizon-fisher-memory (release v1.0)

Code: https://github.com/jeonghoon-ad/finitehorizon-fisher-memory

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• Nonnormal amplification in random balanced neuronal networks
• Improved memory in recurrent neural networks with sequential non-normal dynamics
• Non-normal Recurrent Neural Network (nnRNN): learning long time dependencies while improving expressivity with transient dynamics
• Reversible Recurrent Neural Networks
• On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis
• UnICORNN: A recurrent model for learning very long time dependencies
• Memory Uncertainty Relation and Harmonic Memory in Random Recurrent Networks
• Short-Term Memory in Orthogonal Neural Networks
• How noise affects memory in linear recurrent networks
• Partial Observation of Linear Systems with the Mori-Zwanzig Formalism
• Using Fast Weights to Attend to the Recent Past
• xLSTM: Extended Long Short-Term Memory
• Memory by Design: Bayesian Layers for Sequence Modeling
• Recurrent Kalman Networks: Factorized Inference in High-Dimensional Deep Feature Spaces
• Lindblad-Inspired Multi-Timescale Reservoir Computing with Separable Rotation and Dissipation
• Titans: Learning to Memorize at Test Time
• Duality of Stochastic Observability and Constructability and Links to Fisher Information
• Characterisation of Ces\`aro and $L$-Asymptotic Limits of Matrices
• Asymptotic behaviour of Hilbert space operators with applications

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks