Research map / Probabilistic and scientific modeling / Dynamical systems
Error Forcing in Recurrent Neural Networks
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.
Introduces error forcing, a learning method that nudges RNN activity orthogonally toward the zero-error manifold during training, as an alternative to backpropagation through time.
Problem
Backpropagation through time has known limitations for training RNNs, and it is unclear how to effectively use feedback to shape RNN dynamics during learning.
Approach
Error forcing directly adjusts neural activities during learning, guiding network activity orthogonally toward the zero-error manifold rather than relying solely on gradient propagation through time.
Experimental setup
Builds on: backpropagation through time
Closest papers by meaning
- Feedback control guides credit assignment in recurrent neural networks (NeurIPS 2024 · similarity 0.89)
- Learning Dynamics of RNNs in Closed-Loop Environments (NeurIPS 2025 · similarity 0.86)
- Learning dynamics in linear recurrent neural networks (ICML 2025 · similarity 0.85)
- Dynamical phases of short-term memory mechanisms in RNNs (ICML 2025 · similarity 0.85)
- Partial observation can induce mechanistic mismatches in data-constrained models of neural dynamics (NeurIPS 2024 · similarity 0.85)
- Mechanistic Interpretability of RNNs emulating Hidden Markov Models (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.