Research map / Reinforcement learning / Visual and representation RL
FLEX: an Adaptive Exploration Algorithm for Nonlinear Systems
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 |
|---|---|---|---|
| Use FLEX exploration instead of RHC | Model-based swingup control · Pendulum swingup | observations to solve: 500 -> 50 | Helped |
| Use FLEX exploration instead of RHC | Model-based swingup control · Cartpole swingup | observations to solve: $600$ -> $300$ | Helped |
| Use FLEX exploration instead of RHC | Model-based swingup control · Pendulum swingup | computation time per observation: 8 -> 4 | Helped |
| Use FLEX exploration instead of RHC | Model-based swingup control · Cartpole swingup | computation time per observation: 2 -> 1.6 | Helped |
| Use FLEX instead of random exploration | Coupled damped pendulums · N=2 sample complexity | sample complexity: 20 -> 10 | Helped |
| Use FLEX instead of random exploration | Coupled damped pendulums · N=5 sample complexity | sample complexity: 100 -> 50 | Helped |
| Use FLEX instead of random exploration | Coupled damped pendulums · N=10 sample complexity | sample complexity: 200 -> 100 | Helped |
| Use FLEX instead of random exploration | Coupled damped pendulums · N=20 sample complexity | sample complexity: 500 -> 200 | Helped |
| Use FLEX instead of random exploration | Coupled damped pendulums · N=50 sample complexity | sample complexity: >1000 -> 500 | Helped |
| Use FLEX instead of random exploration | Coupled damped pendulums · N=2 to N=50 | computation time: 1 -> 3; 2 -> 5; 6 -> 10; 60 -> 80; 250 -> 350 | Hurt |
| Replace adaptive FLEX with episodic D-optimal planning | Time-varying force-field exploration · Moving-star force field | tracking dynamics | Hurt |
| Use FLEX rather than random, uniform, or periodic exploration | Nonlinear exploration benchmarks · Pendulum, quadrotor, cartpole, and arm | evaluation error | Helped |
Closest papers by meaning
- Explicit Explore-Exploit Algorithms in Continuous State Spaces (NeurIPS 2019 · similarity 0.91)
- Optimal Exploration for Model-Based RL in Nonlinear Systems (NeurIPS 2023 · similarity 0.91)
- Efficient Exploration in Continuous-time Model-based Reinforcement Learning (NeurIPS 2023 · similarity 0.89)
- OCEAN-MBRL: Offline Conservative Exploration for Model-Based Offline Reinforcement Learning (AAAI 2024 · similarity 0.87)
- LLM-Explorer: A Plug-in Reinforcement Learning Policy Exploration Enhancement Driven by Large Language Models (NeurIPS 2025 · similarity 0.87)
- ASID: Active Exploration for System Identification in Robotic Manipulation (ICLR 2024 · 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.