Engaging Data Scientists in Your Rehabilitation Research

The Restore Center’s scientific challenges will engage data scientists in addressing difficult problems related to using wearable sensor data in a diversity of rehabilitation areas and stimulate long-term collaborations between data scientists and rehabilitation researchers.

Challenge ideas will be solicited from rehabilitation researchers, providing them with the opportunity to have hundreds of data scientists work on a research problem of interest to them. The Restore Center staff will work with the rehabilitation researchers to develop and run the challenges.

Through this approach, we will bring new perspectives to bear on these challenging problems, establish benchmarks to compare approaches, and help identify the state-of-the-art.

Past Scientific Challenges 

Our team has successfully run biomechanics-related challenges at premier machine learning conferences

Learning to Run

NIPS 2017

600+ participants submitted solutions to enable a physiologically-based human model to navigate a complex obstacle course as quickly as possible.

See competition board

Challenge design & outcomes

Top solutions: Read more | See video

AI for Prosthetics

NeurIPS 2018

470+ participants submitted deep reinforcement learning solutions to enable a physiologically-based human model with a prosthetic leg to walk and run based on real-time instructions on movement speed and direction. 

See competition board

Read about top solutions

Learn to Move

NeurIPS 2019

In the Learn to Move: Walk Around competition, 300+ participants submitted deep reinforcement learning solutions to enable a physiologically-based human model to walk and run following velocity commands with minimum effort.

See competition board

Learn about Upcoming Challenges