Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Liangkai Zhou, Susu Xu, Shuqi Zhong, Shan Lin

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.

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

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.
Paper Structure (21 sections, 12 equations, 5 figures, 5 tables, 1 algorithm)

This paper contains 21 sections, 12 equations, 5 figures, 5 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Works
  3. Multi-Task Learning and Multi-Task Causal Discovery
  4. Anti-causal Learning and causal prediction
  5. Urban Event Prediction from Urban Service Data
  6. Problem Definition
  7. Multi-task Structural Causal Model
  8. Multi-task Structural Causal Model
  9. Multi-task Mechanism Module
  10. MAP Anti-Causal Estimation
  11. Training
  12. A Case Study: Urban Event Reconstruction
  13. Evaluation
  14. Datasets
  15. Evaluation Setting
  16. ...and 6 more sections

Figures (5)

  • Figure 1: Multi-task causal graph with shared causal mechanism.
  • Figure 2: MTAC Framework. The white nodes represents parameterized deterministic neural networks, gray nodes represents drawing samples from the respective distribution.
  • Figure 3: Multi-task SEM for mechanism variable $\mathbf{W}_i$. Each block represents an neural network module.
  • Figure 4: Reporting preference model for reporting preference. The SES of residents affect their reporting preference through a set of psychological paths.
  • Figure 5: Subgraph of the learned causal graph representing the generative mechanism of urban events.