arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2202.00531cs.AI
收藏

PRIMA:多任务推理Agent中的规划推理器

PRIMA: Planner-Reasoner Inside a Multi-task Reasoning Agent

Auburn University(奥本大学);Tencent AI Lab(腾讯AI实验室)

Daoming Lyu, Bo Liu, Jianshu Chen

摘要:

我们考虑的问题,多任务推理(MTR),代理可以解决多个任务,通过(一阶)逻辑推理。这种能力对于类人智能来说是必不可少的,因为它具有很强的通用性和处理多个任务的简单性。然而,开发有效的MTR的一个主要挑战是推理能力和效率之间的内在冲突。一个具有MTR能力的代理必须掌握大量的“技能”来处理不同的任务,但是在推理阶段执行特定的任务只需要一小部分直接相关的技能。我们如何保持广泛的推理能力和高效的特定任务性能?为了解决这个问题,我们提出了一个规划推理框架能够最先进的地铁能力和高效率。Reasoner对可共享的(一阶)逻辑推理规则进行建模,Planner从中选择一个子集来组成有效的推理路径。整个模型使用深度强化学习以端到端的方式进行训练,并且在各种领域的实验研究验证了其有效性。

英文摘要:

We consider the problem of multi-task reasoning (MTR), where an agent can solve multiple tasks via (first-order) logic reasoning. This capability is essential for human-like intelligence due to its strong generalizability and simplicity for handling multiple tasks. However, a major challenge in developing effective MTR is the intrinsic conflict between reasoning capability and efficiency. An MTR-capable agent must master a large set of "skills" to tackle diverse tasks, but executing a particular task at the inference stage requires only a small subset of immediately relevant skills. How can we maintain broad reasoning capability and also efficient specific-task performance? To address this problem, we propose a Planner-Reasoner framework capable of state-of-the-art MTR capability and high efficiency. The Reasoner models shareable (first-order) logic deduction rules, from which the Planner selects a subset to compose into efficient reasoning paths. The entire model is trained in an end-to-end manner using deep reinforcement learning, and experimental studies over a variety of domains validate its effectiveness.

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。

↑