• Publications
  • Influence
Learning to select and generalize striking movements in robot table tennis
TLDR
Learning new motor tasks from physical interactions is an important goal for both robotics and machine learning. Expand
  • 308
  • 15
  • PDF
Learning Dynamic Tactile Sensing With Robust Vision-Based Training
TLDR
We propose an efficient approach to infer suitable low-dimensional representations of the tactile data based on only dynamic tactile sensing. Expand
  • 90
  • 8
  • PDF
Weakly-Paired Maximum Covariance Analysis for Multimodal Dimensionality Reduction and Transfer Learning
TLDR
We study the problem of multimodal dimensionality reduction assuming that data samples can be missing at training time, and not all data modalities may be present at application time. Expand
  • 60
  • 8
  • PDF
Interaction primitives for human-robot cooperation tasks
TLDR
In this paper, we propose to learn interaction skills by observing how two humans engage in a similar task. Expand
  • 150
  • 7
  • PDF
Active Reward Learning
TLDR
We propose to learn the reward function through active learning, querying human expert knowledge for a subset of the agent’s rollouts. Expand
  • 71
  • 7
  • PDF
Towards learning hierarchical skills for multi-phase manipulation tasks
TLDR
We present an approach for exploiting the phase structure of tasks in order to learn manipulation skills more efficiently. Expand
  • 80
  • 7
  • PDF
Movement templates for learning of hitting and batting
TLDR
In this paper, we reformulate the Ijspeert framework to incorporate the possibility of specifying a desired hitting point and a desired velocity while maintaining all advantages of the original formulation. Expand
  • 144
  • 6
  • PDF
Probabilistic movement primitives for coordination of multiple human–robot collaborative tasks
TLDR
This paper proposes an interaction learning method for collaborative and assistive robots based on movement primitives based on imitation learning. Expand
  • 93
  • 6
  • PDF
A kernel-based approach to direct action perception
TLDR
We present a non-parametric approach to representing the affordance-bearing subparts of objects. Expand
  • 74
  • 5
  • PDF
Learning to predict phases of manipulation tasks as hidden states
TLDR
We propose a probabilistic model for representing manipulation tasks with multiple phases. Expand
  • 39
  • 4
  • PDF