Regression analysis, the modeling of functional dependencies between variables given noisy data samples, is a central task in many robot learning problems. Relevant problems in robotics include sensor modeling, manipulation learning, learning value functions for control, learning for planning, and many others. Many regression approaches from statistics and machine learning have been proposed to address robotics-related issues such as online updates, estimation of uncertainty, high dimensionality, non-homogeneous noise & smoothness, and missing features.

In this workshop, we would like to develop a common understanding of the benefits and drawbacks of the different regression approaches and to derive practical guidelines for practitioners to choose the right model for solving a given robot learning problem.

As a second focus, we would like to discuss two major points of criticism of standard practices in robot learning research. First, alternative methods are typically evaluated using different metrics and data sets, making it challenging to compare them in a standard way. Second, the use of data-driven and machine learning methods may, in fact, not always be the most suitable approach to solving a given problem, especially when considering decades of established work in model-based control that has been demonstrated to be highly effective. In this respect, we would like to explore what regression problems in robotics can be learnt and to find the subset of the problems that really need to be learnt.

Our goal is to draw researchers from the different communities of robotics, control and machine learning into a discussion of the relevant regression problems to be learnt in robotics. The workshop will address topics such as:

• What are the relevant regression problems in robotics that can be addressed using learning? Which ones of these really need to be learnt?

• What are the most promising and/or evidentially effective regression methods that have been proposed and what are their individual benefits & drawbacks?

• What are the open problems in regression for robotics that still need to be addressed?

• What are common methods for evaluation of regression methods? What metrics or measures are being used for evaluation and, subsequently, which should be used?

• What benchmark regression data sets for robotics are publicly available and can we consolidate them so that comparison of regression methods can be made in a more principled way?

This workshop is part of Robotics: Science and Systems, a recently established international conference series that brings together the leading researchers working on algorithmic and mathematical foundations of robotics, robotics applications, and analysis of robotic systems. Its distinctive features are the highly selective review process and a single track program designed to disseminate the best scientific works in a wide range of robotics-related topics to the public.

From the early conception of the conference, robot learning and function approximation have served as central and significant components of the research contributions presented.