arXiv subjects

Giovanni Varricchione

Publications and source records attributed to Giovanni Varricchione.

3 recordsLinked to original sources

arXiv2025

Pushdown Reward Machines for Reinforcement Learning

Reward machines (RMs) are automata structures that encode (non-Markovian) reward functions for reinforcement learning (RL). RMs can reward any behaviour representable in regular languages and, when paired with RL algorithms that exploit RM structure, have been shown to significantly improve sample efficiency in many domains. In this work, we present pushdown reward machines (pdRMs), an extension of reward machines based on deterministic pushdown automata. pdRMs can recognise and reward temporally extended behaviours representable in deterministic context-free languages, making them more expressive than reward machines. We introduce two variants of pdRM-based policies, one which has access to the entire stack of the pdRM, and one which can only access the top $k$ symbols (for a given constant $k$) of the stack. We propose a procedure to check when the two kinds of policies (for a given environment, pdRM, and constant $k$) achieve the same optimal state values. We then provide theoretical results establishing the expressive power of pdRMs, and space complexity results for the proposed learning problems. Lastly, we propose an approach for off-policy RL algorithms that exploits counterfactual experiences with pdRMs. We conclude by providing experimental results showing how agents can be trained to perform tasks representable in deterministic context-free languages using pdRMs.

cs.AI↗

arXiv2025

From Sound Workflow Nets to LTLf Declarative Specifications by Casting Three Spells

In process management, effective behavior modeling is essential for understanding execution dynamics and identifying potential issues. Two complementary paradigms have emerged in the pursuit of this objective: the imperative approach, representing all allowed runs of a system in a graph-based model, and the declarative one, specifying the rules that a run must not violate in a constraint-based specification. Extensive studies have been conducted on the synergy and comparisons of the two paradigms. To date, though, whether a declarative specification could be systematically derived from an imperative model such that the original behavior was fully preserved (and if so, how) remained an unanswered question. In this paper, we propose a three-fold contribution. (1) We introduce a systematic approach to synthesize declarative process specifications from safe and sound Workflow nets. (2) We prove behavioral equivalence of the input net with the output specification, alongside related guarantees. (3) We experimentally demonstrate the scalability and compactness of our encoding through tests conducted with synthetic and real-world testbeds.

cs.LO↗

arXiv2024

Maximally Permissive Reward Machines

Reward machines allow the definition of rewards for temporally extended tasks and behaviors. Specifying "informative" reward machines can be challenging. One way to address this is to generate reward machines from a high-level abstract description of the learning environment, using techniques such as AI planning. However, previous planning-based approaches generate a reward machine based on a single (sequential or partial-order) plan, and do not allow maximum flexibility to the learning agent. In this paper we propose a new approach to synthesising reward machines which is based on the set of partial order plans for a goal. We prove that learning using such "maximally permissive" reward machines results in higher rewards than learning using RMs based on a single plan. We present experimental results which support our theoretical claims by showing that our approach obtains higher rewards than the single-plan approach in practice.

cs.LG↗