BROWSE
Detail
Conference Paper
Exploring Action Quantization for Data-Driven Learning in Robotic Continuous Control Tasks
Cited 0 time in 
Share 
Share
- Authors
- Ingook Jang, Seonghyun Kim, Samyeul Noh
- Issue Date
- 2024-10
- Citation
- International Conference on Information and Communication Technology Convergence (ICTC) 2024, pp.218-220
- Publisher
- IEEE
- Language
- English
- Type
- Conference Paper
- Abstract
- This paper explores the effectiveness of state-of-the-art action quantization techniques in enhancing offline policy learning for robotic continuous control tasks. We compare three methods - AQuaDem, SAQ, and CEIL - and find that while action quantization improves learning performance, the optimal technique depends on the specific environment and dataset characteristics. Our results highlight the importance of selecting the appropriate method based on these factors.
- KSP Keywords
- Continuous control, Policy learning, data-driven learning, learning performance, specific environment, state-of-The-Art