Timezone: »

 
Learning Representations for Pixel-based Control: What Matters and Why?
Manan Tomar · Utkarsh A Mishra · Amy Zhang · Matthew Taylor

Tue Dec 14 09:00 AM -- 10:00 AM (PST) @
in Ecological Theory of Reinforcement Learning: How Does Task Design Influence Agent Learning? »

Learning representations for pixel-based control has garnered significant attention recently in reinforcement learning. A wide range of methods have been proposed to enable efficient learning, leading to sample complexities similar to those in the full state setting. However, moving beyond carefully curated pixel data sets (centered crop, appropriate lighting, clear background, etc.) remains challenging. In this paper, we adopt a more difficult setting, incorporating background distractors, as a first step towards addressing this challenge. We present a simple baseline approach that can learn meaningful representations with no metric-based learning, no data augmentations, no world-model learning, and no contrastive learning. We then analyze when and why previously proposed methods are likely to fail or reduce to the same performance as the baseline in this harder setting and why we should think carefully about extending such methods beyond the well-curated environments. Our results show that finer categorization of benchmarks on the basis of characteristics like the density of reward, planning horizon of the problem, presence of task-irrelevant components, etc., is crucial in evaluating algorithms. Based on these observations, we propose different metrics to consider when evaluating an algorithm on benchmark tasks. We hope such a data-centric view can motivate researchers to rethink representation learning when investigating how to best apply RL to real-world tasks.

Author Information

Manan Tomar (University of Alberta)
Utkarsh A Mishra (Indian Institute of Technology, Roorkee)
Amy Zhang (FAIR, McGill)
Matthew Taylor (U. of Alberta)

More from the Same Authors