Research map / Probabilistic and scientific modeling / Dynamical systems
Biologically-plausible backpropagation through arbitrary timespans via local neuromodulators
Yuhan Helena Liu, Stephen Smith, Stefan Mihalas, Eric Todd SheaBrown, Uygar Sümbül
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Add filtered modulatory credit propagation across recurrent steps | ModProp · Pattern generation | learning performance | Improved |
| Add filtered modulatory credit propagation across recurrent steps | ModProp · Delayed XOR | learning performance | Improved |
| Add filtered modulatory credit propagation across recurrent steps | ModProp · Sequential MNIST | learning performance | Improved |
| Use time-invariant filtered credit signals instead of precise spatiotemporal credit signals | ModProp · Sequential MNIST | learning performance relative to BPTT | Hurt |
| Replace synapse-specific feedback weights with type-specific feedback weights (ablation) | ModProp · Pattern generation | learning performance | No clear effect |
| Replace synapse-specific feedback weights with type-specific feedback weights (ablation) | ModProp · Delayed XOR | learning performance | No clear effect |
| Restore neuron-specific activation derivatives while retaining temporal averaging (ablation) | ModProp_global · Pattern generation | learning performance | No clear effect |
| Restore neuron-specific activation derivatives while retaining temporal averaging (ablation) | ModProp_global · Delayed XOR | learning performance | No clear effect |
| Restore neuron-specific activation derivatives while retaining temporal averaging (ablation) | ModProp_global · Sequential MNIST | learning performance | No clear effect |
| Use fixed random type-specific modulatory weights instead of weights derived from synaptic weight averages (ablation) | ModProp · Copy task | learning performance | Improved |
| Increase the delayed XOR delay period to 1.5 times the original (ablation) | ModProp with type-specific weights · Delayed XOR | learning performance | Hurt |
Closest papers by meaning
- Biological credit assignment through dynamic inversion of feedforward networks (NeurIPS 2020 · similarity 0.89)
- Volume Transmission Implements Context Factorization to Target Online Credit Assignment and Enable Compositional Generalization (NeurIPS 2025 · similarity 0.88)
- Feedback control guides credit assignment in recurrent neural networks (NeurIPS 2024 · similarity 0.88)
- Structured flexibility in recurrent neural networks via neuromodulation (NeurIPS 2024 · similarity 0.85)
- Discovering plasticity rules that organize and maintain neural circuits (NeurIPS 2024 · similarity 0.85)
- Can Biologically Plausible Temporal Credit Assignment Rules Match BPTT for Neural Similarity? E-prop as an Example (ICML 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.
