analiza aiqbench
This paper introduces TBTCL, a framework that unifies task specification, reward shaping, curriculum design, and policy diagnostics within Temporal Behavior Trees for reinforcement learning. By leveraging the inherent hierarchy of behavior trees, it offers a more expressive alternative to standard formalisms like LTL while enabling automatic extraction of competence profiles and bottleneck adaptation. The approach demonstrates how structural specifications can directly drive robust learning across both discrete and continuous domains.
Temporal Behavior Trees for Reinforcement Learning: Specifications as Rewards, Curricula, and Diagnostics
De la arXiv
Text integralarXiv:2611.13993v1 Announce Type: new Abstract: 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.
Abstract: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.
Machine Learning (cs.LG)
arXiv:2611.13993 [cs.LG]
(or arXiv:2611.13993v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2611.13993 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 274 cuvinte
Subiecte conexe
Towards Surrogate Based Dequantization of Quantum Reinforcement Learning
arXiv · 16 sept. 2026

Reflection’s Beam model hits 501 billion parameters
Quantum Zeitgeist · 5 oct. 2026
Reasoning-Aware Relational Representation Learning for Open-Vocabulary Scene Graph Generation
arXiv · 11 nov. 2026
Q-Learning for Reachability in MEC-Free MDPs
arXiv · 2 oct. 2026