Transfer Learning and Curriculum Learning in Sokoban

2021-05-25 · Zhao Yang, Mike Preuss, Aske Plaat

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 more complex tasks, but not vice versa. We have also studied which part of the knowledge is most important for transfer to succeed, and identify which layers should be used for pre-training.

📄 PDF Abstract BibTeX arXiv:2105.11702

Code (0)

등록된 구현이 없습니다.

Tasks

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)SokobanTransfer Learning

A Novel Automated Curriculum Strategy to Solve Hard Sokoban Planning Instances

2021-10-03 · NeurIPS 2020 12 · Dieqiao Feng, Carla P. Gomes, Bart Selman

In recent years, we have witnessed tremendous progress in deep reinforcement learning (RL) for tasks such as Go, Chess, video games, and robot control. Nevertheless, other combinatorial domains, such as AI planning, stil…

Deep Reinforcement Learning Reinforcement Learning (RL)Sokoban

Solving Hard AI Planning Instances Using Curriculum-Driven Deep Reinforcement Learning

2020-06-04 · Dieqiao Feng, Carla P. Gomes, Bart Selman

Despite significant progress in general AI planning, certain domains remain out of reach of current AI planning systems. Sokoban is a PSPACE-complete planning task and represents one of the hardest domains for current AI…

Deep Reinforcement Learning reinforcement-learning Reinforcement Learning (RL)Sokoban

Zero-Shot Transfer with Deictic Object-Oriented Representation in Reinforcement Learning

2018-12-01 · NeurIPS 2018 12 · Ofir Marom, Benjamin Rosman

Object-oriented representations in reinforcement learning have shown promise in transfer learning, with previous research introducing a propositional object-oriented framework that has provably efficient learning bounds …

Object reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)+2

Understanding Curriculum Learning in Large Language Models via Cross-Difficulty Optimization Dynamics

2026-08-18 · Zhikai Ding, Ziyi Ye arxiv

Curriculum learning has been widely adopted in the post-training of large language models by organizing training data from easy to hard. However, its effectiveness varies substantially across reasoning tasks, suggesting …

AI in Game Playing: Sokoban Solver

2018-06-29 · Anand Venkatesan, Atishay Jain, Rakesh Grewal

Artificial Intelligence is becoming instrumental in a variety of applications. Games serve as a good breeding ground for trying and testing these algorithms in a sandbox with simpler constraints in comparison to real lif…

AI Agent Sokoban