Optimal transport (OT) is a powerful geometric tool used to compare and ...
The co-adaptation of robots has been a long-standing research endeavour ...
Physics simulators have shown great promise for conveniently learning
re...
We present a data-efficient framework for solving sequential decision-ma...
The framework of Simulation-to-real learning, i.e, learning policies in
...
In recent years, domain randomization has gained a lot of traction as a
...
Sample-efficient domain adaptation is an open problem in robotics. In th...
Domain adaptation is a common problem in robotics, with applications suc...
Few-shot adaptation is a challenging problem in the context of
simulatio...
Modern reinforcement learning methods suffer from low sample efficiency ...
Training end-to-end deep robot policies requires a lot of domain-, task-...