Research map / Reinforcement learning / Visual and representation RL
Multi-Sacle Dynamic Coding Improved Spiking Actor Network for Reinforcement Learning
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 |
|---|---|---|---|
| Replace the DAN actor with MDC-SAN using population coding and dynamic neurons | TD3 actor · Ant-v3 | max average reward: 4671 -> 5628 | Helped |
| Replace the DAN actor with MDC-SAN using population coding and dynamic neurons | TD3 actor · HalfCheetah-v3 | max average reward: 10020 -> 11657 | Helped |
| Replace the DAN actor with MDC-SAN using population coding and dynamic neurons | TD3 actor · Walker2d-v3 | max average reward: 4139 -> 5566 | Helped |
| Replace the DAN actor with MDC-SAN using population coding and dynamic neurons | TD3 actor · Hopper-v3 | max average reward: 3396 -> 3636 | Helped |
| Add population coding to a DAN actor (ablation) | DAN · Ant-v3, HalfCheetah-v3, Walker2d-v3, and Hopper-v3 | average reward | Mixed |
| Add population coding to a spiking actor network (ablation) | SAN · Ant-v3, HalfCheetah-v3, Walker2d-v3, and Hopper-v3 | average reward | Helped |
| Use population coding alone instead of Poisson rate coding (ablation) | SAN with dynamic neurons · Ant-v3, HalfCheetah-v3, Walker2d-v3, and Hopper-v3 | average reward | Helped |
| Add uniform, Poisson, or deterministic rate coding after population coding instead of using population-coded analog inputs directly (ablation) | SAN with dynamic neurons · Ant-v3, HalfCheetah-v3, Walker2d-v3, and Hopper-v3 | average reward | Hurt |
| Use dynamic neurons instead of LIF neurons (ablation) | SAN with population coding · Ant-v3, HalfCheetah-v3, Walker2d-v3, and Hopper-v3 | average reward | Helped |
Closest papers by meaning
- GRSN: Gated Recurrent Spiking Neurons for POMDPs and MARL (AAAI 2025 · similarity 0.89)
- Adaptive Surrogate Gradients for Sequential Reinforcement Learning in Spiking Neural Networks (NeurIPS 2025 · similarity 0.87)
- Proxy Target: Bridging the Gap Between Discrete Spiking Neural Networks and Continuous Control (NeurIPS 2025 · similarity 0.86)
- Gated Attention Coding for Training High-Performance and Efficient Spiking Neural Networks (AAAI 2024 · similarity 0.86)
- SpikingSSMs: Learning Long Sequences with Sparse and Parallel Spiking State Space Models (AAAI 2025 · similarity 0.86)
- Decision SpikeFormer: Spike-Driven Transformer for Decision Making (CVPR 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.