Research map / Probabilistic and scientific modeling / Dynamical systems
No Free Lunch from Deep Learning in Neuroscience: A Case Study through Models of the Entorhinal-Hippocampal Circuit
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Cartesian or polar position targets instead of tuned DoS bump readouts (ablation) | path-integrating RNNs · grid cell emergence | fraction of networks with grid-like units | Hurt |
| use Gaussian readouts instead of DoS readouts (ablation) | path-integrating RNNs · grid cell emergence | grid score | Hurt |
| use tuned Difference-of-Softmaxes readouts (ablation) | path-integrating RNNs · grid cell emergence | grid score | Helped |
| use equi-norm DoG readouts instead of DoS (ablation) | ReLU RNNs · grid cell emergence | grid score | Hurt |
| add non-negativity (ReLU) constraint with DoG readouts in feasible region (ablation) | ReLU RNNs · grid cell emergence | grid score: 1 of 1086 runs matched DoS | No clear effect |
| add place-cell-like heterogeneity (multi-field, multi-scale) to DoS readouts (ablation) | best DoS RNNs · grid cell emergence | grid score | Hurt |
| add place-cell-like heterogeneity to DoS readouts (ablation) | best DoS RNNs · path integration | position decoding error | No clear effect |
| modestly vary DoS width sigma_E (ablation) | DoS RNNs · grid cell emergence | grid score | Hurt |
| sweep DoS width sigma_E (ablation) | DoS RNNs · grid period | grid period | Mixed |
| remove dropout from Gaussian-readout network (ablation) | Gaussian readout RNN · grid cell emergence | grid-like units | Hurt |
Closest papers by meaning
- A Multi-Region Brain Model to Elucidate the Role of Hippocampus in Spatially Embedded Decision-Making (ICML 2025 · similarity 0.88)
- The Expressive Leaky Memory Neuron: an Efficient and Expressive Phenomenological Neuron Model Can Solve Long-Horizon Tasks. (ICLR 2024 · similarity 0.88)
- A simple model of co-emergence of grid and place fields (NeurIPS 2026 · similarity 0.88)
- Scaling Laws and Tradeoffs in Recurrent Networks of Expressive Neurons (NeurIPS 2026 · similarity 0.87)
- Neuroformer: Multimodal and Multitask Generative Pretraining for Brain Data (ICLR 2024 · similarity 0.87)
- A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks (NeurIPS 2025 · similarity 0.86)
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.