Abstract:We introduce Temporal Ensemble Logic (TEL), a monadic, first-order modal logic for linear-time temporal reasoning. TEL includes primitive temporal constructs such as ``always up to $t$ time later'' ($\Box_t$), ``sometimes before $t$ time in the future'' ($\Diamond_t$), and ``$t$-time later'' $\varphi_t$. TEL has been motivated from the requirement for rigor and reproducibility for cohort specification and discovery in clinical and population health research, to fill a gap in formalizing temporal reasoning in biomedicine. Existing logical frameworks such as linear temporal logic are too restrictive to express temporal and sequential properties in biomedicine, or too permissive in semantic constructs, such as in Halpern-Shoham logic, to serve this purpose. In this paper, we first introduce TEL in a general set up, with discrete and dense time as special cases. We then focus on the theoretical development of discrete TEL on the temporal domain of positive integers $\mathbb{N}^+$, denoted as ${\rm TEL}_{\mathbb{N}^+}$. ${\rm TEL}_{\mathbb{N}^+}$ is strictly more expressive than the standard monadic second order logic, characterized by Büchi automata. We present its formal semantics, a proof system, and provide a proof for the undecidability of the satisfiability of ${\rm TEL}_{\mathbb{N}^+}$. We also include initial results on expressiveness and decidability fragments for ${\rm TEL}_{\mathbb{N}^+}$, followed by application outlook and discussions.

A Flexible and Efficient Temporal Logic Tool for Python: PyTeLo

Metric Temporal Equilibrium Logic over Timed Traces

Standpoint Linear Temporal Logic

Temporal Ensemble Logic

Robust, Expressive, and Quantitative Linear Temporal Logics: Pick any Two for Free (full version)

Robust, Expressive, and Quantitative Linear Temporal Logics: Pick any Two for Free

An Efficient Formula Synthesis Method with Past Signal Temporal Logic

Runtime Verification of Spatio-Temporal Specification Language

NL2STL: Transformation from Logic Natural Language to Signal Temporal Logics Using Llama2

Synthesizing Efficiently Monitorable Formulas in Metric Temporal Logic

TILP: Differentiable Learning of Temporal Logical Rules on Knowledge Graphs

Temporal Stream Logic: Synthesis Beyond the Bools

DeepSTL -- From English Requirements to Signal Temporal Logic

Mining Interpretable Spatio-temporal Logic Properties for Spatially Distributed Systems

TLINet: Differentiable Neural Network Temporal Logic Inference

The Temporal Logic Synthesis Format TLSF v1.2

Adaptive Teaching of Temporal Logic Formulas to Learners with Preferences

PyReason: Software for Open World Temporal Logic

Scalable Anytime Algorithms for Learning Fragments of Linear Temporal Logic

Formalizing MLTL Formula Progression in Isabelle/HOL

PSY-TaLiRo: A Python Toolbox for Search-Based Test Generation for Cyber-Physical Systems