PaperlayerView exact revision on arXiv

Source-linked AI summary

Resilient Concurrent Causal Discovery for Topological Event Sequences

Jiyu Tian, Junhao Dong, Mingchu Li, Lingling Fang, Liming Chen, Andreas Holzinger, Zheng Yan, Yew Soon Ong

arXiv:2608.21815v1cs.LG

TL;DR

Existing causal-discovery methods struggle with concurrent causal relationships and incomplete event sequences in telecommunication networks. RCCD combines influence-aware hyperedge causal attention with mask-based alternating optimization, and experiments on synthetic and real-world datasets report improved accuracy and robustness over existing methods.

Abstract

from arXiv · show

Causal discovery on topological event sequences is crucial for ensuring the reliability of networks. However, existing methods struggle to capture the complex causal relationships arising from concurrent events and lack robustness to incomplete event sequences. To address these issues, we propose a resilient concurrent causal discovery method, termed RCCD, enabling robust learning of causal graphs from topological event sequences. Specifically, we first introduce an influence-aware hyperedge causal attention mechanism, which incorporates event duration into the embedding representation, aggregates concurrent event features via hyperedge causal convolution, and injects network prior knowledge to capture the complex many-to-one causal interactions. Furthermore, we design a masked-based alternating causal optimization framework, which forces the model to recover masked event types based on context through self-supervised mask reconstruction, thereby enhancing the resilience of the predictor to missing data. To validate the effectiveness of our method, we conduct extensive experiments on both simulated and real-world telecommunication network datasets. Experimental results demonstrate that the proposed method significantly outperforms existing state-of-the-art methods in both accuracy and robustness, making it more suitable for real-world telecommunication network environments.

I. INTRODUCTION

Telecommunication event sequences encode topology-constrained, concurrent, duration-dependent alarms, yet missing observations challenge causal discovery. RCCD addresses these challenges with topology-aware attention and mask-based alternating optimization, and experiments evaluate its accuracy and robustness.

  • Telecommunication alarms form chronological event sequences that can help locate root causes, block chain reactions, and trace anomaly propagation.Source: §I · p. 1
  • Concurrent alarms, duration effects, heterogeneous devices, topology constraints, and missing records create complex causal-learning conditions.Source: §I · p. 1, Figure 1 · p. 1
  • Existing methods struggle to model collaborative concurrent causality and generally assume complete event sequences.Source: §I · p. 2
  • RCCD introduces influence-aware hyperedge causal attention that encodes duration, aggregates concurrent events, and incorporates topology priors.Source: §I · p. 2
  • RCCD uses mask-based alternating causal optimization with self-supervised reconstruction to improve prediction robustness under incomplete observations.Source: §I · p. 2
  • Experiments on synthetic and real-world datasets assess RCCD's causal-learning performance and robustness against existing methods.Source: §I · p. 2

II. RELATED WORK

Related work spans topology-agnostic and topology-aware causal discovery, including statistical and neural point-process methods. Despite advances in temporal and causal attention, existing approaches incompletely address missing events, many-to-one causality, topology priors, and concurrent relationships.

A. Causal Learning on Event Sequences

B. Temporal Causal Attention Mechanism

III. PROBLEM FORMULATION

The paper formulates causal discovery for telecommunication event sequences using network topology and seeks a binary causal graph, then presents RCCD's topology-aware, duration- and concurrency-sensitive architecture with masked optimization.

III. PROBLEM FORMULATION

  • The network is modeled as an undirected graph whose physical connections define topological relationships among nodes.Source: §III · p. 4
  • Event types are represented by a directed acyclic graph and binary causal matrix, where each matrix entry indicates whether one type causes another.Source: §III · p. 4
  • Each event records timestamp, type, topological node, and nonnegative duration within a sequence of length L.Source: §III · p. 4
  • The stated goal is to infer the causal relationships among event types from observed events and network topology, producing the final binary causal matrix.Source: §III · p. 4

A. Overview

  • RCCD combines duration-, concurrency-, and topology-aware attention with masked multi-step optimization to discover a causal graph robustly from non-i.i.d. sequences.Source: A. Overview · p. 4
  • IHCA encodes duration, distinguishes instantiated events, and aggregates consecutive local events into hyperedge representations for concurrent causal modeling.Source: A. Overview · p. 4
  • MACO masks historical events during predictor training, reconstructs their types from context, and alternately optimizes predictor and causal parameters.Source: A. Overview · p. 4
  • The overall pipeline combines event and topology processing, hyperedge causal attention, causal and k-hop constraints, topology decay, and alternating optimization.Source: A. Overview · pp. 4–5, Figure 2 · p. 5

B. Influence-aware Hyperedge Causal Attention Mechanism

IHCA enriches event representations with duration information alongside event type and occurrence time, while suppressing duration features for instantaneous events.

1) Event Influence-Aware Encoding:

2) Hyperedge Causal Attention Score:

The hyperedge causal attention mechanism replaces isolated pairwise attention with causal convolution over local event histories and applies causal and topological priors to constrain information flow.

2) Hyperedge Causal Attention Score:

3) Topology-Constrained Causal Attention Mechanism:

2) Hyperedge Causal Attention Score:

C. Mask-based Alternation Causal Optimization

MACO improves resilience to incomplete event sequences by masking eligible historical events during predictor training and reconstructing their types from context within alternating causal optimization.

C. Mask-based Alternation Causal Optimization

1) Mask Generation Strategy:

2) EM-Style Two-Stage Alternating Optimization Framework:

RCCD uses an EM-style alternating framework that trains a predictor for event and masked-event reconstruction, then optimizes causal graph parameters with the predictor fixed. The four-task objective is intended to improve causal discovery stability under missing data.

3) Causal Judgment:

After alternating training, RCCD converts learned causal probabilities into a binary causal graph using a threshold. The experiments cover stationary causal learning, incomplete observations, hyperparameters, ablations, and a real-world case study.

  • RCCD declares a causal edge from vi to vj when the learned causal probability ω(Ãi,j) exceeds threshold ρ.Source: 3) Causal Judgment · p. 8
  • Topology decay is incorporated into attention weights so RCCD jointly learns causal probabilities and topological structure.Source: 3) Causal Judgment · p. 8
  • The evaluation includes baseline comparisons, incomplete-observation robustness, hyperparameter analysis, ablations, and a real-world causal-graph case study.Source: §V · p. 8
  • Synthetic datasets vary event types, physical devices, causal strength, and event counts, with ground-truth causal and topology graphs generated for benchmarking.Source: A. Dataset · p. 8
  • The real-world microwave datasets contain 24 or 25 event types, 439 or 474 nodes, and 64,599 or 48,573 event records.Source: A. Dataset · p. 8

B. Baselines

RCCD is compared with classical, neural, point-process, topology-aware, and large-language-model baselines using standard causal-graph metrics. It achieves strong F1 and SHD performance across synthetic and real-world datasets.

Baselines

  • Baselines include PC, GraNDAG, CORL, THPs, TNPAR, S2GCSL, CausalNET, and TCCD, alongside DeepSeek-v3.2, Gemini-v3.0, and ChatGPT-v5.1.Source: B. Baselines · pp. 8–9

Evaluation Criteria

  • Causal graphs are evaluated with Precision, Recall, F1 Score, and Structural Hamming Distance.SHD counts the minimum edge insertions, deletions, or reversals needed to transform the predicted graph into the ground truth.Source: C. Evaluation Criteria · p. 9

Stable datasets

Large language models

F. Performance Comparison on Observing Incomplete Sequences

RCCD maintains leading causal-discovery performance as event-missing rates increase from 5% to 20%. Its F1 Score declines modestly and its SHD remains comparatively stable across synthetic and real telecommunication datasets.

Incomplete observations

F1 Score

Overall robustness

Structural Hamming Distance

G. Hyperparameters Sensitivity Analysis

RCCD maintains stable causal-discovery performance across tested regularizer settings, with minimal fluctuations on real-world data. Its complementary IHCA and MACO components improve performance under complete and incomplete observations.

Sparsity regularizer σ1

  • On synthetic datasets, changing σ1 from 0.1 to 0.3 leaves precision, recall, F1 Score, and SHD unchanged.On the 24V-439N-Microwave dataset, F1 changes from 0.5313 to 0.5274 across the same σ1 range.Source: G. Hyperparameters Sensitivity Analysis · p. 11

Acyclicity regularizer σ2

  • For σ2 on 15V-30N-Synthetic, F1 remains 0.8421 and SHD remains 3 as σ2 increases from 0.1 to 0.3.On 20V-40N-Synthetic, F1 changes from 0.8750 to 0.8235 and SHD from 4 to 6 between σ2 = 0.1 and 0.2.Source: G. Hyperparameters Sensitivity Analysis · p. 11

Hyperparameter sensitivity

  • RCCD maintains stable high performance across tested sparsity and acyclicity regularizer settings, indicating low hyperparameter sensitivity.The evaluation varies σ1 and σ2 from 0.1 to 0.3 across synthetic and real-world datasets.Source: G. Hyperparameters Sensitivity Analysis · pp. 11–12

Ablation study

  • IHCA mainly improves recall and F1, whereas MACO helps preserve structural accuracy and precision under incomplete observations.The paper attributes the optimal performance across test conditions to combining both components.Source: H. Ablation Study · p. 13

VI. DISCUSSION

RCCD combines duration- and concurrency-aware causal attention with masked alternating optimization to learn causal graphs from incomplete event sequences. Experiments report improved causal discovery accuracy and robustness, while the authors identify nonrandom missingness and baseline-reproduction differences as limitations.

Method

  • IHCA incorporates event duration and aggregates local concurrent events into hyper-edges to capture jointly triggered causal relationships.The mechanism uses a nonlinear gated mapping for duration and causal convolution for local concurrent-event aggregation.Source: A. Why does the RCCD work? · p. 13, §VII · p. 13

Limitations

  • The masking strategy uses predefined random probabilities, which may not adapt fully to nonrandom event missingness in telecommunications networks.The authors propose learning the masking distribution from realistic collection noise in future work.Source: A. Why does the RCCD work? · p. 13
  • Reproduction differences in text interpretation or parameter tuning may affect the fairness of comparisons for methods without released source code.TNPAR and TCCD were reproduced from their paper descriptions rather than official implementations.Source: B. Threats to Validity · p. 13

Method

  • MACO improves robustness to missing events through self-supervised reconstruction within a mask-based alternating optimization framework.The framework enhances predictor robustness to incomplete observations while jointly optimizing causal structure learning.Source: §VII · p. 13

Empirical effectiveness

  • RCCD significantly outperforms existing state-of-the-art methods in discovery performance and robustness across synthetic and real datasets.The reported evaluation covers multiple datasets and incomplete-observation settings.Source: §VII · p. 13

Related papers

Loading 2608.21815v1…