# From Video Game to Real Robot: The Transfer between Action Spaces

## TL;DR

This paper explores how to transfer policies learned in video games to real robots, focusing on learning new action spaces with minimal retraining, achieving high success rates in both simulation and real-world tests.

## Contribution

It demonstrates that partial neural network retraining enables effective transfer of action spaces from video games to robots, reducing training time and effort.

## Key findings

- Achieved above 90% success rate in simulation
- Successfully transferred policies to real robots
- Learned new action spaces autonomously

## Abstract

Deep reinforcement learning has proven to be successful for learning tasks in simulated environments, but applying same techniques for robots in real-world domain is more challenging, as they require hours of training. To address this, transfer learning can be used to train the policy first in a simulated environment and then transfer it to physical agent. As the simulation never matches reality perfectly, the physics, visuals and action spaces by necessity differ between these environments to some degree. In this work, we study how general video games can be directly used instead of fine-tuned simulations for the sim-to-real transfer. Especially, we study how the agent can learn the new action space autonomously, when the game actions do not match the robot actions. Our results show that the different action space can be learned by re-training only part of neural network and we obtain above 90% mean success rate in simulation and robot experiments.