Learning of feature points without additional supervision improves reinforcement learning from images
This addresses the challenge of simplifying feature extraction in reinforcement learning from images, though it is incremental as it builds on existing feature point methods.
The paper tackled the problem of learning feature point representations for vision-based control without requiring unsupervised pre-training or additional supervision, and achieved performance competitive with state-of-the-art methods on DeepMind Control Suite tasks.
In many control problems that include vision, optimal controls can be inferred from the location of the objects in the scene. This information can be represented using feature points, which is a list of spatial locations in learned feature maps of an input image. Previous works show that feature points learned using unsupervised pre-training or human supervision can provide good features for control tasks. In this paper, we show that it is possible to learn efficient feature point representations end-to-end, without the need for unsupervised pre-training, decoders, or additional losses. Our proposed architecture consists of a differentiable feature point extractor that feeds the coordinates of the estimated feature points directly to a soft actor-critic agent. The proposed algorithm yields performance competitive to the state-of-the art on DeepMind Control Suite tasks.