Research map / Reinforcement learning / Policy gradient and actor-critic
Theoretically Principled Deep RL Acceleration via Nearest Neighbor Function Approximation
Junhong Shen, Lin F. Yang
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 the soft nearest-neighbor update module | TD3 · Ant | average return: 3425.07 -> 4727.44 | Improved |
| Add the soft nearest-neighbor update module | DDPG · Ant | average return: 1025.82 -> 1115.00 | Improved |
| Add the soft nearest-neighbor update module | TD3 · Hopper | average return: 2708.25 -> 3704.40 | Improved |
| Add the soft nearest-neighbor update module | DDPG · Hopper | average return: 2029.65 -> 3202.64 | Improved |
| Add the soft nearest-neighbor update module | TD3 · Walker2d | average return: 4260.23 -> 5170.20 | Improved |
| Add the soft nearest-neighbor update module | DDPG · Walker2d | average return: 2633.86 -> 2917.25 | Improved |
| Add the soft nearest-neighbor update module | TD3 · HalfCheetah | average return: 7045.53 -> 8341.74 | Improved |
| Add the soft nearest-neighbor update module | DDPG · HalfCheetah | average return: 6385.43 -> 7001.52 | Improved |
| Use L2 distance for image similarity instead of a learned distance metric (ablation) | NNAC · CartPole-v1 pixel input | learning stability | Hurt |
| Use a learned distance metric for pixel input (ablation) | NNAC · CartPole-v1 pixel input | sample efficiency | Mixed |
| Use projected states rather than intrinsic state descriptors (ablation) | NNAC · CartPole-v1 | convergence steps | No clear effect |
Closest papers by meaning
- Stop Regressing: Training Value Functions via Classification for Scalable Deep RL (ICML 2024 · similarity 0.87)
- The Effective Horizon Explains Deep RL Performance in Stochastic Environments (ICLR 2024 · similarity 0.87)
- SRL: Scaling Distributed Reinforcement Learning to Over Ten Thousand Cores (ICLR 2024 · similarity 0.86)
- SimBa: Simplicity Bias for Scaling Up Parameters in Deep Reinforcement Learning (ICLR 2025 · similarity 0.86)
- Adaptive $Q$-Network: On-the-fly Target Selection for Deep Reinforcement Learning (ICLR 2025 · similarity 0.86)
- Exploration-Driven Policy Optimization in RLHF: Theoretical Insights on Efficient Data Utilization (ICML 2024 · 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.
