Research map / Architectures and efficiency / RNNs and normalization
Unfolding the Black Box of Recurrent Neural Networks for Path Integration
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.
Trains recurrent neural networks on a path integration task and applies neuroscience-inspired analysis methods to unfold how the trained RNN computes position.
Problem
How the neural system accomplishes path integration, a computation essential for spatial navigation, remains unresolved despite experimental identification of neural correlates.
Approach
Uses RNNs trained on a path integration task as a model system and borrows neuroscience prior knowledge and methods to open the network's black box and explain its internal computation.
Closest papers by meaning
- Not so griddy: Internal representations of RNNs path integrating more than one agent (NeurIPS 2024 · similarity 0.89)
- Global Distortions from Local Rewards: Neural Coding Strategies in Path-Integrating Neural Systems (NeurIPS 2024 · similarity 0.88)
- A simple model of co-emergence of grid and place fields (NeurIPS 2026 · similarity 0.86)
- On Conformal Isometry of Grid Cells: Learning Distance-Preserving Position Embedding (ICLR 2025 · similarity 0.86)
- The motion planning neural circuit in goal-directed navigation as Lie group operator search (NeurIPS 2024 · similarity 0.85)
- Flexible mapping of abstract domains by grid cells via self-supervised extraction and projection of generalized velocity signals (NeurIPS 2024 · 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.