Multi-Task Anti-Causal Learning for Reconstructing Urban Events from Residents' Reports

2026-03-12

AI Summary

This study addresses the multi-task anti-causal inference problem of reverse-engineering the underlying causes of urban incidents—such as illegal parking, abandoned properties, and sanitation issues—from resident reports. To this end, the authors propose the MTAC framework, which introduces cross-task causal invariance into anti-causal learning for the first time. MTAC decouples generative mechanisms via a shared backbone and task-specific heads, jointly modeling structured multi-task structural equation models. It further integrates causal discovery with maximum a posteriori (MAP) inference to optimize latent mechanism variables and cause strengths. Experiments on real-world data from Manhattan and Newark demonstrate that the method reduces mean absolute error (MAE) by up to 34.61% compared to strong baselines, significantly improving both accuracy and generalization in reconstructing the root causes of urban events.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Many real-world machine learning tasks are anti-causal: they require inferring latent causes from observed effects. In practice, we often face multiple related tasks where part of the forward causal mechanism is invariant across tasks, while other components are task-specific. We propose Multi-Task Anti-Causal learning (MTAC), a framework for estimating causes from outcomes and confounders by explicitly exploiting such cross-task invariances. MTAC first performs causal discovery to learn a shared causal graph and then instantiates a structured multi-task structural equation model (SEM) that factorizes the outcome-generation process into (i) a task-invariant mechanism and (ii) task-specific mechanisms via a shared backbone with task-specific heads. Building on the learned forward model, MTAC performs maximum A posteriori (MAP)based inference to reconstruct causes by jointly optimizing latent mechanism variables and cause magnitudes under the learned causal structure. We evaluate MTAC on the application of urban event reconstruction from resident reports, spanning three tasks:parking violations, abandoned properties, and unsanitary conditions. On real-world data collected from Manhattan and the city of Newark, MTAC consistently improves reconstruction accuracy over strong baselines, achieving up to 34.61\% MAE reduction and demonstrating the benefit of learning transferable causal mechanisms across tasks.

Problem

• anti-causal learning
• multi-task learning
• urban event reconstruction
• causal discovery
• latent cause inference

Innovation

• Multi-Task Anti-Causal Learning
• Causal Discovery
• Structural Equation Model
• Invariant Mechanism
• MAP Inference