不依赖搜索或人类知识的解谜：一种非自然语言方法
Puzzle Solving without Search or Human Knowledge: An Unnatural Language Approach
摘要：
应用生成式预训练Transformer（GPT-2）来学习文本存档的游戏符号，为探索稀疏奖励游戏提供了一个模型环境。Transformer架构被证明可以在描述迷宫、魔方和数独求解器的已解决文本档案上进行训练。该方法受益于微调的Transformer架构可视化合理的战略以外的任何指导，从人类的经验或领域的专业知识。游戏的大搜索空间（$>10^{19}$）提供了一个谜题环境，其中解决方案几乎没有中间奖励和解决挑战的最后一步。
英文摘要：
The application of Generative Pre-trained Transformer (GPT-2) to learn text-archived game notation provides a model environment for exploring sparse reward gameplay. The transformer architecture proves amenable to training on solved text archives describing mazes, Rubik's Cube, and Sudoku solvers. The method benefits from fine-tuning the transformer architecture to visualize plausible strategies derived outside any guidance from human heuristics or domain expertise. The large search space ($>10^{19}$) for the games provides a puzzle environment in which the solution has few intermediate rewards and a final move that solves the challenge.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
