Insights into Learning Broadcast Protocols: (Short Paper)

Dana Fisman; Noa Izsak; Swen Jacobs

doi:10.1007/978-3-032-10759-6_20

Insights into Learning Broadcast Protocols: (Short Paper)

Dana Fisman
, Noa Izsak
, Swen Jacobs

Institute for the Theory of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Broadcast protocols (BPs) are a formal model of distributed systems with an unbounded number of processes communicating through broadcasts. We study the problem of passively learning BPs from execution traces, focusing on the class of fine BPs which does not have hidden states and admits a cutoff. We present a passive learning algorithm with a constraint-based approach that guarantees consistency with the sample, and returns a minimal equivalent BP if the sample is sufficiently complete (i.e., subsumes a characteristic set). Furthermore, we describe LeoParDS, the first tool that implements these techniques, supporting the practical inference of fine BPs, as well as tasks that include sample generation and approximate equivalence checking. This work was previously published at AAAI’24 [8] and later implemented at ATVA’24 [12]. We summarize its main results here to foster discussion within the cybersecurity and verification community. This short paper is intended as a concise overview for readers unfamiliar with both prior publications.

Original language	English
Title of host publication	Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings
Editors	Adi Akavia, Shlomi Dolev, Anna Lysyanskaya, Rami Puzis
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	306-313
Number of pages	8
ISBN (Print)	9783032107589
DOIs	https://doi.org/10.1007/978-3-032-10759-6_20
State	Published - 1 Jan 2026
Event	9th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2025 - Be'er Sheva, Israel Duration: 4 Dec 2025 → 5 Dec 2025

Publication series

Name	Lecture Notes in Computer Science
Volume	16244 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	9th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2025
Country/Territory	Israel
City	Be'er Sheva
Period	4/12/25 → 5/12/25

Keywords

Broadcast Protocols
Learning Theory
Multiagent Systems

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-032-10759-6_20

Cite this

Fisman, D., Izsak, N., & Jacobs, S. (2026). Insights into Learning Broadcast Protocols: (Short Paper). In A. Akavia, S. Dolev, A. Lysyanskaya, & R. Puzis (Eds.), Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings (pp. 306-313). (Lecture Notes in Computer Science; Vol. 16244 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-032-10759-6_20

Fisman, Dana ; Izsak, Noa ; Jacobs, Swen. / Insights into Learning Broadcast Protocols : (Short Paper). Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings. editor / Adi Akavia ; Shlomi Dolev ; Anna Lysyanskaya ; Rami Puzis. Springer Science and Business Media Deutschland GmbH, 2026. pp. 306-313 (Lecture Notes in Computer Science).

@inproceedings{691e88535dcc48849c9244c5be451bd1,

title = "Insights into Learning Broadcast Protocols: (Short Paper)",

abstract = "Broadcast protocols (BPs) are a formal model of distributed systems with an unbounded number of processes communicating through broadcasts. We study the problem of passively learning BPs from execution traces, focusing on the class of fine BPs which does not have hidden states and admits a cutoff. We present a passive learning algorithm with a constraint-based approach that guarantees consistency with the sample, and returns a minimal equivalent BP if the sample is sufficiently complete (i.e., subsumes a characteristic set). Furthermore, we describe LeoParDS, the first tool that implements these techniques, supporting the practical inference of fine BPs, as well as tasks that include sample generation and approximate equivalence checking. This work was previously published at AAAI{\textquoteright}24 [8] and later implemented at ATVA{\textquoteright}24 [12]. We summarize its main results here to foster discussion within the cybersecurity and verification community. This short paper is intended as a concise overview for readers unfamiliar with both prior publications.",

keywords = "Broadcast Protocols, Learning Theory, Multiagent Systems",

author = "Dana Fisman and Noa Izsak and Swen Jacobs",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.; 9th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2025 ; Conference date: 04-12-2025 Through 05-12-2025",

year = "2026",

month = jan,

day = "1",

doi = "10.1007/978-3-032-10759-6\_20",

language = "English",

isbn = "9783032107589",

series = "Lecture Notes in Computer Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "306--313",

editor = "Adi Akavia and Shlomi Dolev and Anna Lysyanskaya and Rami Puzis",

booktitle = "Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings",

address = "Germany",

}

Fisman, D, Izsak, N & Jacobs, S 2026, Insights into Learning Broadcast Protocols: (Short Paper). in A Akavia, S Dolev, A Lysyanskaya & R Puzis (eds), Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings. Lecture Notes in Computer Science, vol. 16244 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 306-313, 9th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2025, Be'er Sheva, Israel, 4/12/25. https://doi.org/10.1007/978-3-032-10759-6_20

Insights into Learning Broadcast Protocols: (Short Paper). / Fisman, Dana; Izsak, Noa; Jacobs, Swen.
Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings. ed. / Adi Akavia; Shlomi Dolev; Anna Lysyanskaya; Rami Puzis. Springer Science and Business Media Deutschland GmbH, 2026. p. 306-313 (Lecture Notes in Computer Science; Vol. 16244 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Insights into Learning Broadcast Protocols

T2 - 9th International Symposium on Cyber Security, Cryptology, and Machine Learning, CSCML 2025

AU - Fisman, Dana

AU - Izsak, Noa

AU - Jacobs, Swen

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

PY - 2026/1/1

Y1 - 2026/1/1

N2 - Broadcast protocols (BPs) are a formal model of distributed systems with an unbounded number of processes communicating through broadcasts. We study the problem of passively learning BPs from execution traces, focusing on the class of fine BPs which does not have hidden states and admits a cutoff. We present a passive learning algorithm with a constraint-based approach that guarantees consistency with the sample, and returns a minimal equivalent BP if the sample is sufficiently complete (i.e., subsumes a characteristic set). Furthermore, we describe LeoParDS, the first tool that implements these techniques, supporting the practical inference of fine BPs, as well as tasks that include sample generation and approximate equivalence checking. This work was previously published at AAAI’24 [8] and later implemented at ATVA’24 [12]. We summarize its main results here to foster discussion within the cybersecurity and verification community. This short paper is intended as a concise overview for readers unfamiliar with both prior publications.

AB - Broadcast protocols (BPs) are a formal model of distributed systems with an unbounded number of processes communicating through broadcasts. We study the problem of passively learning BPs from execution traces, focusing on the class of fine BPs which does not have hidden states and admits a cutoff. We present a passive learning algorithm with a constraint-based approach that guarantees consistency with the sample, and returns a minimal equivalent BP if the sample is sufficiently complete (i.e., subsumes a characteristic set). Furthermore, we describe LeoParDS, the first tool that implements these techniques, supporting the practical inference of fine BPs, as well as tasks that include sample generation and approximate equivalence checking. This work was previously published at AAAI’24 [8] and later implemented at ATVA’24 [12]. We summarize its main results here to foster discussion within the cybersecurity and verification community. This short paper is intended as a concise overview for readers unfamiliar with both prior publications.

KW - Broadcast Protocols

KW - Learning Theory

KW - Multiagent Systems

UR - https://www.scopus.com/pages/publications/105023438893

U2 - 10.1007/978-3-032-10759-6_20

DO - 10.1007/978-3-032-10759-6_20

M3 - Conference contribution

AN - SCOPUS:105023438893

SN - 9783032107589

T3 - Lecture Notes in Computer Science

SP - 306

EP - 313

BT - Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings

A2 - Akavia, Adi

A2 - Dolev, Shlomi

A2 - Lysyanskaya, Anna

A2 - Puzis, Rami

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 4 December 2025 through 5 December 2025

ER -

Fisman D, Izsak N, Jacobs S. Insights into Learning Broadcast Protocols: (Short Paper). In Akavia A, Dolev S, Lysyanskaya A, Puzis R, editors, Cyber Security, Cryptology, and Machine Learning - 9th International Symposium, CSCML 2025, Proceedings. Springer Science and Business Media Deutschland GmbH. 2026. p. 306-313. (Lecture Notes in Computer Science). doi: 10.1007/978-3-032-10759-6_20

Insights into Learning Broadcast Protocols: (Short Paper)

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this