Jens Schreiter

About

Jens Schreiter is a leading researcher in machine learning and robotics, whose work focuses on making Gaussian process (GP) regression computationally efficient for real-world applications. His major contribution lies in developing sparse GP models that enable fast, scalable regression on large datasets without sacrificing accuracy. His most influential work, "Efficient sparsification for Gaussian process regression" (2016, 32 citations), introduced novel techniques for selecting representative subsets of training data, dramatically reducing computational costs. This breakthrough was directly applied in his highly cited paper "Sparse Gaussian process regression for compliant, real-time robot control" (2015, 23 citations), where he demonstrated how these efficient GP models can power adaptive, safe physical interactions between robots and humans. Schreiter also advanced the field with "Fast greedy insertion and deletion in sparse Gaussian process regression" (2015), which introduced a streamlined criterion for dynamically updating GP models—a critical capability for autonomous systems operating in changing environments. His work bridges the gap between theoretical machine learning and practical robotics, making him a key figure in the development of data-efficient, real-time learning systems for compliant robot control.

Research Focus

Key Achievements

Top Papers

1. 1
   Efficient sparsification for Gaussian process regression

   32 citations · 2016
2. 2
   Sparse Gaussian process regression for compliant, real-time robot control

   23 citations · 2015
3. 3
   Fast greedy insertion and deletion in sparse Gaussian process regression.

   3 citations · 2015

Key Collaborators

Contact & Links