Beyond Steps RL

Beyond Steps RL symbolizes the ambition to go beyond conventional boundaries in reinforcement learning (RL). The name reflects the integration of cutting-edge RL techniques with robotic locomotion, especially focusing on quadrupedal robots. It emphasizes innovation, exploration, and the pursuit of advancements that push RL applications beyond mere movement—toward solving real-world challenges with precision and adaptability.

Study Records

No.	Paper	Presenter	Date	File
1	Not Only Rewards but Also Constraints	Jihong Kim	24.09.01	paper_review
2	Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion	JungYeon Lee	24.09.01	paper_review
3	Spinning Up in Deep RL	Chanwoo Park	24.09.22	paper_review
4	Not Only Rewards but Also Constraints II	Jehee Lee	24.10.06	paper_review
5	Constrained Policy Optimization	Jinwon Kim	24.10.06	paper_review
6	IPO: Interior-point Policy Optimization under Constraints	JungYeon Lee	24.10.06	paper_review
7	TRPO/PPO	Chanwoo Park	24.10.20	paper_review
8	Constrained Policy Optimization II	Jihong Kim	24.10.20	paper_review
9	Learning-based legged locomotion; state of the art and future perspectives	Jinwon Kim	24.11.17	paper_review
10	pympc-quadruped	Jihong Kim	25.01.05	code_review
11	Model Predictive Control	JungYeon Lee	25.01.05	paper_review

Members

JungYeon Lee	Jihong Kim	Jinwon Kim	Chanwoo Park