# Curiosity-driven Exploration by Self-supervised Prediction

## TL;DR

This paper introduces a curiosity-driven exploration method using self-supervised prediction errors in visual feature space, enabling efficient learning in environments with sparse or no extrinsic rewards, and improving generalization to new scenarios.

## Contribution

It proposes a scalable curiosity mechanism based on prediction errors in learned feature space, addressing high-dimensional environments and environmental irrelevance.

## Key findings

- Curiosity significantly reduces the number of interactions needed to reach goals.
- Agents explore more efficiently without extrinsic rewards.
- Knowledge transfer accelerates exploration in unseen environments.

## Abstract

In many real-world scenarios, rewards extrinsic to the agent are extremely sparse, or absent altogether. In such cases, curiosity can serve as an intrinsic reward signal to enable the agent to explore its environment and learn skills that might be useful later in its life. We formulate curiosity as the error in an agent's ability to predict the consequence of its own actions in a visual feature space learned by a self-supervised inverse dynamics model. Our formulation scales to high-dimensional continuous state spaces like images, bypasses the difficulties of directly predicting pixels, and, critically, ignores the aspects of the environment that cannot affect the agent. The proposed approach is evaluated in two environments: VizDoom and Super Mario Bros. Three broad settings are investigated: 1) sparse extrinsic reward, where curiosity allows for far fewer interactions with the environment to reach the goal; 2) exploration with no extrinsic reward, where curiosity pushes the agent to explore more efficiently; and 3) generalization to unseen scenarios (e.g. new levels of the same game) where the knowledge gained from earlier experience helps the agent explore new places much faster than starting from scratch. Demo video and code available at https://pathak22.github.io/noreward-rl/