Skip to main content Skip to main navigation


Learning probabilistic discriminative models of grasp affordances under limited supervision

Ayse Erkan; Oliver Kroemer; Renaud Detry; Yasemin Altun; Justus H. Piater; Jan Peters
In: 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS-2010), October 18-22, Taipei, Taiwan, Province of China, Pages 1586-1591, IEEE, 2010.


This paper addresses the problem of learning and efficiently representing discriminative probabilistic models of object-specific grasp affordances particularly when the number of labeled grasps is extremely limited. The proposed method does not require an explicit 3D model but rather learns an implicit manifold on which it defines a probability distribution over grasp affordances. We obtain hypothetical grasp configurations from visual descriptors that are associated with the contours of an object. While these hypothetical configurations are abundant, labeled configurations are very scarce as these are acquired via time-costly experiments carried out by the robot. Kernel logistic regression (KLR) via joint kernel maps is trained to map the hypothesis space of grasps into continuous class-conditional probability values indicating their achievability. We propose a soft-supervised extension of KLR and a framework to combine the merits of semi-supervised and active learning approaches to tackle the scarcity of labeled grasps. Experimental evaluation shows that combining active and semi-supervised learning is favorable in the existence of an oracle. Furthermore, semi-supervised learning outperforms supervised learning, particularly when the labeled data is very limited.

Weitere Links