An Analysis of the Impact of Dataset Characteristics on Data-Driven Reinforcement Learning for a Robotic Long-Horizon Task
- Resource Type
- Conference
- Authors
- Jang, Ingook; Noh, Samyeul; Kim, Seonghyun; Lee, Donghun
- Source
- 2023 14th International Conference on Information and Communication Technology Convergence (ICTC) Information and Communication Technology Convergence (ICTC), 2023 14th International Conference on. :1681-1683 Oct, 2023
- Subject
- Bioengineering
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Fields, Waves and Electromagnetics
Power, Energy and Industry Applications
Signal Processing and Analysis
Transportation
Training
Costs
Reinforcement learning
Manipulators
Data models
Safety
Information and communication technology
Data-driven reinforcement learning
robot manipulation
long-horizon task
- Language
- ISSN
- 2162-1241
Data-driven reinforcement learning (RL) is a cost-effective method for training agents without online interaction with the real-world environment. This approach involves collecting and storing data from various sources such as expert demonstrations or random policies, and learning from these datasets without further online interaction with the environment. However, learning an optimal behavioral model from offline data is challenging as it may not cover the entire state-action space. The paper discusses an experimental study analyzing how dataset characteristics impact the performance on a long-horizon robot manipulation task using a robotic arm. The goal of the paper is to provide guidance on strategically organizing datasets for training agents via data-driven RL.