The control of continuous-time environments while actively deciding when to take costly
observations in time is a crucial yet unexplored problem, particularly relevant to real-world …

We investigate the use of transformer sequence models as dynamics models (TDMs) for
control. We find that TDMs exhibit strong generalization capabilities to unseen …

Designing physical artifacts that serve a purpose---such as tools and other functional
structures---is central to engineering as well as everyday human behavior. Though …

Deep learning has been widely used within learning algorithms for robotics. One
disadvantage of deep networks is that these networks are black-box representations …

Model-based reinforcement learning (MBRL) is a sample efficient technique to obtain control
policies, yet unavoidable modeling errors often lead performance deterioration. The model …

State of the art reinforcement learning has enabled training agents on tasks of ever
increasing complexity. However, the current paradigm tends to favor training agents from …

Many real-world offline reinforcement learning (RL) problems involve continuous-time
environments with delays. Such environments are characterized by two distinctive features …

Monte Carlo methods have become increasingly relevant for control of non-differentiable
systems, approximate dynamics models, and learning from data. These methods scale to …

Reinforcement learning (RL) is able to solve complex sequential decision-making tasks but
is currently limited by sample efficiency and required computation. To improve sample …

We consider the Imitation Learning (IL) setup where expert data are not collected on the
actual deployment environment but on a different version. To address the resulting …