ridm@nrct.go.th   ระบบคลังข้อมูลงานวิจัยไทย   รายการโปรดที่คุณเลือกไว้

Using Dimensionality Reduction to Exploit Constraints in Reinforcement Learning

หน่วยงาน Edinburgh Research Archive, United Kingdom

รายละเอียด

ชื่อเรื่อง : Using Dimensionality Reduction to Exploit Constraints in Reinforcement Learning
นักวิจัย : Bitzer, Sebastian , Howard, Matthew , Vijayakumar, Sethu
คำค้น : Informatics , Computer Science , Robotics
หน่วยงาน : Edinburgh Research Archive, United Kingdom
ผู้ร่วมงาน : -
ปีพิมพ์ : 2553
อ้างอิง : http://hdl.handle.net/1842/3644
ที่มา : -
ความเชี่ยวชาญ : -
ความสัมพันธ์ : -
ขอบเขตของเนื้อหา : -
บทคัดย่อ/คำอธิบาย :

Reinforcement learning in the high-dimensional, continuous spaces typical in robotics, remains a challenging problem. To overcome this challenge, a popular approach has been to use demonstrations to find an appropriate initialisation of the policy in an attempt to reduce the number of iterations needed to find a solution. Here, we present an alternative way to incorporate prior knowledge from demonstrations of individual postures into learning, by extracting the inherent problem structure to find an efficient state representation. In particular, we use probabilistic, nonlinear dimensionality reduction to capture latent constraints present in the data. By learning policies in the learnt latent space, we are able to solve the planning problem in a reduced space that automatically satisfies task constraints. As shown in our experiments, this reduces the exploration needed and greatly accelerates the learning. We demonstrate our approach for learning a bimanual reaching task on the 19-DOF KHR-1HV humanoid.

บรรณานุกรม :
Bitzer, Sebastian , Howard, Matthew , Vijayakumar, Sethu . (2553). Using Dimensionality Reduction to Exploit Constraints in Reinforcement Learning.
    กรุงเทพมหานคร : Edinburgh Research Archive, United Kingdom .
Bitzer, Sebastian , Howard, Matthew , Vijayakumar, Sethu . 2553. "Using Dimensionality Reduction to Exploit Constraints in Reinforcement Learning".
    กรุงเทพมหานคร : Edinburgh Research Archive, United Kingdom .
Bitzer, Sebastian , Howard, Matthew , Vijayakumar, Sethu . "Using Dimensionality Reduction to Exploit Constraints in Reinforcement Learning."
    กรุงเทพมหานคร : Edinburgh Research Archive, United Kingdom , 2553. Print.
Bitzer, Sebastian , Howard, Matthew , Vijayakumar, Sethu . Using Dimensionality Reduction to Exploit Constraints in Reinforcement Learning. กรุงเทพมหานคร : Edinburgh Research Archive, United Kingdom ; 2553.