Log In Sign Up

GridToPix: Training Embodied Agents with Minimal Supervision

by   Unnat Jain, et al.

While deep reinforcement learning (RL) promises freedom from hand-labeled data, great successes, especially for Embodied AI, require significant work to create supervision via carefully shaped rewards. Indeed, without shaped rewards, i.e., with only terminal rewards, present-day Embodied AI results degrade significantly across Embodied AI problems from single-agent Habitat-based PointGoal Navigation (SPL drops from 55 to 0) and two-agent AI2-THOR-based Furniture Moving (success drops from 58 Google Football-based 3 vs. 1 with Keeper (game score drops from 0.6 to 0.1). As training from shaped rewards doesn't scale to more realistic tasks, the community needs to improve the success of training with terminal rewards. For this we propose GridToPix: 1) train agents with terminal rewards in gridworlds that generically mirror Embodied AI environments, i.e., they are independent of the task; 2) distill the learned policy into agents that reside in complex visual worlds. Despite learning from only terminal rewards with identical models and RL algorithms, GridToPix significantly improves results across tasks: from PointGoal Navigation (SPL improves from 0 to 64) and Furniture Moving (success improves from 1 improves from 0.1 to 0.6). GridToPix even helps to improve the results of shaped reward training.


Handling Sparse Rewards in Reinforcement Learning Using Model Predictive Control

Reinforcement learning (RL) has recently proven great success in various...

Off-Policy Reinforcement Learning with Delayed Rewards

We study deep reinforcement learning (RL) algorithms with delayed reward...

Towards Empathic Deep Q-Learning

As reinforcement learning (RL) scales to solve increasingly complex task...

A Study on Dense and Sparse (Visual) Rewards in Robot Policy Learning

Deep Reinforcement Learning (DRL) is a promising approach for teaching r...

Happiness Pursuit: Personality Learning in a Society of Agents

Modeling personality is a challenging problem with applications spanning...

Distributed Ensembles of Reinforcement Learning Agents for Electricity Control

Deep Reinforcement Learning (or just "RL") is gaining popularity for ind...

Policy Learning Using Weak Supervision

Most existing policy learning solutions require the learning agents to r...