Research map / Reinforcement learning / Visual and representation RL
Deep Reinforcement Learning with Plasticity Injection
Evgenii Nikishin, Junhyuk Oh, Georg Ostrovski, Clare Lyle, Razvan Pascanu, Will Dabney, Andre Barreto
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 plasticity injection during training | Double DQN · Atari 57 | aggregate score: 20% | Improved |
| Start with a smaller network and add plasticity injection at 50M frames | Double DQN · Atari games | IQM | No clear effect |
| Use a smaller network up to 50M frames, then inject plasticity | Double DQN · Atari games | wallclock time: 20 hours | Improved |
| Add plasticity injection | Double DQN · Phoenix | return | Improved |
| Add plasticity injection | Double DQN · Assault | return | No clear effect |
| Add plasticity injection | Double DQN · Robotank | return | No clear effect |
| Apply injection to the entire network instead of sharing the encoder (ablation) | Plasticity injection · Atari 57 | IQM | Hurt |
| Copy the encoder at injection instead of sharing it (ablation) | Plasticity injection · Atari 57 | IQM | Mixed |
| Keep the original parameter set unfrozen (ablation) | Plasticity injection · Atari 57 | IQM | No clear effect |
| Add further injections at 100M and 150M frames (ablation) | Plasticity injection · Atari 57, standard network | IQM | No clear effect |
| Add further injection rounds (ablation) | Plasticity injection with smaller networks · Atari 57 | IQM | Improved |
| Omit the frozen random head subtraction that corrects initial outputs (ablation) | Plasticity injection · Yars Revenge | return | Hurt |
Closest papers by meaning
- A Study of Plasticity Loss in On-Policy Deep Reinforcement Learning (NeurIPS 2024 · similarity 0.91)
- Revisiting Plasticity in Visual Reinforcement Learning: Data, Modules and Training Stages (ICLR 2024 · similarity 0.90)
- Adaptive Rational Activations to Boost Deep Reinforcement Learning (ICLR 2024 · similarity 0.88)
- Mitigating Plasticity Loss in Continual Reinforcement Learning by Reducing Churn (ICML 2025 · similarity 0.88)
- Stay Hungry, Keep Learning: Sustainable Plasticity for Deep Reinforcement Learning (ICML 2025 · similarity 0.87)
- Neuroplastic Expansion in Deep Reinforcement Learning (ICLR 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.
