TBTCL 框架用 Temporal Behavior Trees 统一强化学习的规格、奖励与课程设计
Temporal Behavior Trees for Reinforcement Learning: Specifications as Rewards, Curricula, and Diagnostics
用行为树一个东西同时搞定任务描述、稠密奖励、课程和策略诊断，不用再在 LTL 和奖励函数之间来回翻译。做机器人 RL 或形式化方法的可以看看论文里的课程 DAG 提取算法，6 个规格的实验覆盖了离散和连续环境。
arXiv 论文提出 TBTCL（Temporal Behavior Tree Centric Learning），把 Temporal Behavior Trees（TBT）作为规格引导强化学习的统一接口。传统做法中任务定义、奖励塑形、课程设计和策略诊断分属不同数学对象，需要额外转换步骤，且丢失规格中已有的子任务边界、顺序约束等信息。
TBT 在该框架中同时承担四个角色：定义任务、基于鲁棒性语义生成稠密奖励、自动提取课程 DAG、并通过轨迹分割构建可理解的能力画像，还能自动把奖励调整到瓶颈子任务上。论文称 TBT 的表达力强于 LTL 等常用规格语言，且天然具有层次结构。
作者在离散与连续两类域的 6 个规格上验证了 TBTCL 的效果。对做机器人任务规整或形式化方法结合 RL 的研究者来说，这篇提供了一套可复用的算法：从 TBT 语义直接推导稠密奖励和课程，省去手工设计流程。
Temporal Behavior Trees for Reinforcement Learning: Specifications as Rewards, Curricula, and Diagnostics
Reinforcement learning from formal specifications often separates task definition, reward shaping, curriculum design, and policy diagnostics into different mathematical objects. This fragmentation requires additional conversion steps and discards structure already present in the specification, such as subtask boundaries, ordering constraints, and quantitative measures. We introduce TBTCL (Temporal Behavior Tree Centric Learning), a framework that uses Temporal Behavior Trees (TBTs) as a unified interface for specification-guided reinforcement learning. A single TBT specifies the task, induces dense robustness-based rewards, defines a curriculum, and supports better understanding of learned policies. By adapting TBT syntax and semantics for specification-guided reinforcement learning, we provide a formalism that is more expressive than popular specification languages such as LTL and inherently hierarchical. We present algorithms to automatically extract a curriculum DAG and derive dense, robustness-based rewards directly from TBT semantics. Furthermore, we utilize TBT trace segmentation to build understandable competence profiles and automatically adapt the reward toward bottlenecks. We evaluate TBTCL on six specifications across discrete and continuous domains and show that Temporal Behavior Trees can serve as a specification, a reward, a curriculum, and a diagnostic.
本条情报归属 AITOP「论文」核心追踪栏目。在跨区域产业观察坐标中，该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：TBTCL、Temporal Behavior Trees、强化学习、LTL，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明：本文原始线索采编自「arXiv cs.LG」（@Vladimir Krsmanovic, Bernd Finkbeiner），经 AITOP 多维评分管线结构化提炼。