Transfer Learning and Curriculum Learning in Sokoban
Zhao Yang, Mike Preuss, Aske Plaat

TL;DR
This paper demonstrates how transfer and curriculum learning techniques can significantly accelerate reinforcement learning in Sokoban by reusing learned features and structuring tasks from simple to complex.
Contribution
It introduces a method for applying transfer learning and curriculum learning to Sokoban, showing how feature reuse and task progression improve learning efficiency.
Findings
Reusing feature representations accelerates learning in complex Sokoban instances.
Curriculum learning from simple to complex tasks enhances transfer effectiveness.
Pre-trained layers are crucial for successful transfer in reinforcement learning.
Abstract
Transfer learning can speed up training in machine learning and is regularly used in classification tasks. It reuses prior knowledge from other tasks to pre-train networks for new tasks. In reinforcement learning, learning actions for a behavior policy that can be applied to new environments is still a challenge, especially for tasks that involve much planning. Sokoban is a challenging puzzle game. It has been used widely as a benchmark in planning-based reinforcement learning. In this paper, we show how prior knowledge improves learning in Sokoban tasks. We find that reusing feature representations learned previously can accelerate learning new, more complex, instances. In effect, we show how curriculum learning, from simple to complex tasks, works in Sokoban. Furthermore, feature representations learned in simpler instances are more general, and thus lead to positive transfers towards…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsReinforcement Learning in Robotics · Artificial Intelligence in Games · Topic Modeling
