Research map / Probabilistic and scientific modeling / Dynamical systems
Efficient Training of Minimal and Maximal Low-Rank 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.
Proposes methods for efficiently fitting low-rank RNNs in teacher-training settings by building on the neural engineering framework (NEF), covering both minimal and maximal rank variants.
Problem
Fitting and interpreting low-rank recurrent neural networks used as models of neural computation in cognitive tasks remains an open problem; existing fitting is costly.
Approach
Builds on the neural engineering framework (NEF), in which RNNs are constructed from low-rank connectivity, to fit low-rank RNNs efficiently in teacher-training settings, for both minimal and maximal rank cases.
Experimental setup
Builds on: neural engineering framework
Closest papers by meaning
- Inferring stochastic low-rank recurrent neural networks from neural data (NeurIPS 2024 · similarity 0.92)
- The interplay between randomness and structure during learning in RNNs (NeurIPS 2020 · similarity 0.90)
- Extracting computational mechanisms from neural data using low-rank RNNs (NeurIPS 2022 · similarity 0.89)
- Dynamical Isometry and a Mean Field Theory of RNNs: Gating Enables Signal Propagation in Recurrent Neural Networks (ICML 2018 · similarity 0.89)
- Learning Reveals Invisible Structure in Low-Rank RNNs (NeurIPS 2026 · similarity 0.89)
- Recurrent neural networks: vanishing and exploding gradients are not the end of the story (NeurIPS 2024 · similarity 0.89)
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.