Mar 15, 2021 · Our proposed method utilizes training data collected entirely offline in the real-world which removes the need of intensive online explorations ...

Jun 4, 2021 · Our proposed method utilizes training data collected entirely offline in the real-world which removes the need of intensive online explorations ...

Learning robust driving policies without online exploration · D. Graves, Nhat M. Nguyen, +2 authors. Jun Luo · Published in IEEE International Conference… 15 ...

We propose a multi-time-scale predictive representation learning method to efficiently learn robust driving policies in an offline manner that generalize ...

We propose a multi-time-scale predictive representation learning method to efficiently learn robust driving policies in an offline manner that generalize ...

The paper proposes a multi-time-scale predictive representation learning method to efficiently learn robust driving policies in an offline manner without online ...

In this work, we aim to achieve efficient end-to-end learning of driving policies in dynamic multi-agent environments. Predicting and anticipating future events ...

Learning Robust Driving Policies ... (without on-policy data). Ohn-Bar, Prakash, Behl ... Exploring Data Aggregation in Policy Learning for Vision-based Urban ...

There are two key challenges to employing the large amounts of unconstrained and unlabeled online data for training robust vision-based navigation policies.

We present a novel robust reinforcement learning approach with safety guarantees to attain trustworthy decision-making for autonomous vehicles.