Symmetric Machine Theory of Mind
Symmetric Machine Theory of Mind
复制标题
DOI：
--
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Melanie Sclar;Graham Neubig;Yonatan Bisk
中科院分区：
文献类型：
--
作者：
Melanie Sclar;Graham Neubig;Yonatan Bisk
心理理论，即对他人的想法和欲望进行建模的能力，是人类社会智力的基石。这使得这对机器学习社区来说是一个重要的挑战，但以前的工作主要是尝试设计代理，将其他人的“心理状态”建模为被动观察者或特定预定义的角色，例如在说话人-听话人场景中。相反，我们建议在更一般的对称场景中对机器心理理论进行建模。我们引入了一个多代理环境SymmToM，在这个环境中，像在现实生活中一样，所有代理都可以说话、倾听、看到其他代理，并在世界上自由移动。最大限度提高代理人报酬的有效策略需要它发展一种心理理论。我们表明，与没有这种心理理论模型的代理相比，模拟他人心理状态的强化学习代理的性能有了显着的提高。重要的是，我们最好的代理仍然无法获得与其他代理的黄金标准心理状态相媲美的性能，这表明在多代理场景中对心理理论进行建模是一个非常开放的挑战。代码可以在https：//githeb.com/msclar/symmtom上找到。
Theory of mind, the ability to model others’ thoughts and desires, is a cornerstone of human social intelligence. This makes it an important challenge for the machine learning community, but previous works mainly attempt to design agents that model the “mental state” of others as passive observers or in specific predefined roles, such as in speaker-listener scenarios. In contrast, we propose to model machine theory of mind in a more general symmetric scenario. We introduce a multi-agent environment SymmToM where, like in real life, all agents can speak, listen, see other agents, and move freely through the world. Effective strategies to maximize an agent’s reward require it to develop a theory of mind. We show that reinforcement learning agents that model the mental states of others achieve significant performance improvements over agents with no such theory of mind model. Importantly, our best agents still fail to achieve performance comparable to agents with access to the gold-standard mental state of other agents, demonstrating that the modeling of theory of mind in multi-agent scenarios is very much an open challenge. Code can be found at https: //github.com/msclar/symmtom.