User Tools

Site Tools


6a.053.UVA - Assistive Agents for Self-Represented Litigants

Project - Summary

In America today fewer than one in five low-income individuals have access to the legal help they need.

Many individuals become self-represented litigants (SRLs) and achieve poor outcomes as a result.

Federal spending on legal aid would need to increase from $300M to $1.6B to resolve these issues.

The goal of this project is to identify innovative algorithmic and analytic methods for meeting these needs without increasing federal spending.

Specifically, we aim to provide helpful and personalized assistance to SRLs via optimized matching of SRLs with attorneys and courts.

Project - Team

Team Member Role Email Phone Number Academic Site/IAB
Matthew Gerber PI (434) 924-5397 University of Virginia
Mark Rucker Student (479) 434-0164 University of Virginia
Dan Becker Project Mentor Phone: (651) 848-4655 Mobile: (651) 503-1264 Thomson Reuters

Project - Deliverables

1 Working implementation of inverse reinforcement learning projection algorithm
2 Forward reinforcement learning algorithm to discover optimal policy
3 Construction of agent-based model from data traces, IRL and RL algorithms
4 Identification of underlying goals behind observed behavior patterns

Project - Presentation Video

Project - Documents

FilenameFilesizeLast modified
6a.053.uva_assistive_agents_for_self-represented_litigants_poster_2017_fall_meeting.pptx284.0 KiB2019/08/14 15:37
6a.053.uva_cvdi_mid-year_report.docx236.1 KiB2019/08/14 15:37
6a.053.uva_confluence_project_page.pdf144.7 KiB2019/08/14 15:37
6a.053.uva_quad_chart.pdf226.2 KiB2019/08/14 15:37
6a.053.uva_poster_pdf.pdf352.2 KiB2019/08/14 15:37
6a.053.uva_executive_summary.docx51.9 KiB2019/08/14 15:37
6a.053.uva_ppt_presentation.pptx57.7 KiB2019/08/14 15:37
projects/year6/6a.053.uva.txt · Last modified: 2019/08/14 16:25 by sally.johnson