Malte Viering

About

Malte Viering is a researcher at the forefront of robot learning, with a primary focus on the challenge of reward function design—a critical bottleneck in applying reinforcement learning to real-world robotic tasks. His work centers on *active reward learning*, a paradigm that enables robots to efficiently query human feedback to infer desired behaviors without requiring hand-crafted reward signals. In his seminal 2014 paper, "Active Reward Learning," which has garnered 88 citations, Viering demonstrated how robots can actively select informative queries to learn reward functions for tasks like grasping, where objective success measures are often unavailable. He extended this framework in his 2015 work, "Active reward learning with a novel acquisition function" (50 citations), introducing a more efficient query selection strategy that reduces the number of human interactions needed. These contributions have been instrumental in making robot learning more sample-efficient and practical for real-world deployment. By addressing the fundamental problem of reward specification, Viering's research has influenced subsequent work in interactive robot learning and human-robot collaboration, establishing him as a key contributor to the field's progress toward autonomous systems that can learn from limited human guidance.

Research Focus

Key Achievements

Top Papers

1. 1
   Active Reward Learning

   88 citations · 2014
2. 2
   Active reward learning with a novel acquisition function

   50 citations · 2015

Key Collaborators

Contact & Links