Auxiliary Tasks Speed Up Learning PointGoal Navigation
This work addresses the computational inefficiency in embodied AI navigation, making training more practical, though it is incremental as it builds on existing methods with auxiliary tasks.
The paper tackles the problem of slow sample and time efficiency in learning PointGoal Navigation, a task requiring agents to navigate to a point in unseen environments, by using self-supervised auxiliary tasks with attention-based combination, resulting in a 5.5x faster learning speed to match previous state-of-the-art performance and a 0.16 SPL improvement at 40M frames.
PointGoal Navigation is an embodied task that requires agents to navigate to a specified point in an unseen environment. Wijmans et al. showed that this task is solvable but their method is computationally prohibitive, requiring 2.5 billion frames and 180 GPU-days. In this work, we develop a method to significantly increase sample and time efficiency in learning PointNav using self-supervised auxiliary tasks (e.g. predicting the action taken between two egocentric observations, predicting the distance between two observations from a trajectory,etc.).We find that naively combining multiple auxiliary tasks improves sample efficiency,but only provides marginal gains beyond a point. To overcome this, we use attention to combine representations learnt from individual auxiliary tasks. Our best agent is 5.5x faster to reach the performance of the previous state-of-the-art, DD-PPO, at 40M frames, and improves on DD-PPO's performance at 40M frames by 0.16 SPL. Our code is publicly available at https://github.com/joel99/habitat-pointnav-aux.