Active Geospatial Search for Efficient Tenant Eviction Outreach

TL;DR

This work tackles the problem of improving tenant eviction outreach by identifying at-risk properties within a fixed canvassing budget. It introduces Active Geospatial Search (AGS), a budgeted sequential decision framework that combines multimodal parcel features with a reinforcement learning-based search policy, aiming to maximize detected eviction targets. To scale to large urban areas, it presents Hierarchical AGS (HAGS), which decomposes the region into regions and parcels, sharing a prediction module across levels and training region-level and region-specific policies with REINFORCE and supervised learning. Experiments on eviction data from a large city show that AGS and especially HAGS outperform strong baselines by substantial margins, highlighting the potential to increase timely access to legal aid, financial assistance, and other supports for tenants at risk of eviction.

Abstract

Tenant evictions threaten housing stability and are a major concern for many cities. An open question concerns whether data-driven methods enhance outreach programs that target at-risk tenants to mitigate their risk of eviction. We propose a novel active geospatial search (AGS) modeling framework for this problem. AGS integrates property-level information in a search policy that identifies a sequence of rental units to canvas to both determine their eviction risk and provide support if needed. We propose a hierarchical reinforcement learning approach to learn a search policy for AGS that scales to large urban areas containing thousands of parcels, balancing exploration and exploitation and accounting for travel costs and a budget constraint. Crucially, the search policy adapts online to newly discovered information about evictions. Evaluation using eviction data for a large urban area demonstrates that the proposed framework and algorithmic approach are considerably more effective at sequentially identifying eviction cases than baseline methods.

Figures (8)

• Figure 1: Policy network architecture.
• Figure 2: HAGS policy network architecture.
• Figure 3: Multi-Modal Feature Extraction module for each Parcel.
• Figure 4: Performance Analysis Across Different Evaluation Set.
• Figure 5: Performance Analysis Across Different Evaluation Set.