Research map / Architectures and efficiency / RNNs and normalization
Differentiable plasticity: training plastic neural networks with backpropagation
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 |
|---|---|---|---|
| Add trainable plastic connections to the recurrent network | Recurrent network · Binary pattern memorization (1{,}000-bit patterns) | bit error: <1\% | Helped |
| Add trainable plastic connections instead of using an LSTM | 50-bit binary pattern memorization · Binary pattern memorization (50-bit patterns) | episodes to reach mean error below .01: ~2{,}000 vs. 500{,}000 episodes | Helped |
| Use independent plasticity coefficients for each connection rather than one shared coefficient (ablation) | Differentiable plasticity network · Natural image memorization | reconstruction performance | Helped |
| Use trainable plastic connections for the output layer | Convolutional Omniglot classifier · Omniglot 5-way 1-shot | accuracy: 98.3\% $\pm$ 0.80 | Helped |
| Use individually learned plasticity coefficients instead of no plasticity | Recurrent maze agent · Maze exploration task | reward | Helped |
| Use one shared learnable plasticity coefficient instead of individually learned coefficients (ablation) | Recurrent maze agent · Maze exploration task | reward | Hurt |
Closest papers by meaning
- Learning the Plasticity: Plasticity-Driven Learning Framework in Spiking Neural Networks (NeurIPS 2025 · similarity 0.91)
- Slow and Steady Wins the Race: Maintaining Plasticity with Hare and Tortoise Networks (ICML 2024 · similarity 0.88)
- Model Based Inference of Synaptic Plasticity Rules (NeurIPS 2024 · similarity 0.87)
- Rethinking the Stability-Plasticity Trade-off in Continual Learning from an Architectural Perspective (ICML 2025 · similarity 0.87)
- Self-Normalized Resets for Plasticity in Continual Learning (ICLR 2025 · similarity 0.87)
- Neuroplastic Expansion in Deep Reinforcement Learning (ICLR 2025 · similarity 0.87)
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.