SYS 4581/6581: AI for Social Good (AI4G)

Overview

Spring Semester Undergraduate and Graduate Course at UVA

AI for Social Good challenges students to learn and apply Artificial Intelligence techniques to social and global causes and to see its impact in real terms. Students will learn about different AI topics and algorithms and work on a project to apply those methods to develop solutions and applications for social and global good. We will cover the three essential elements of an AI system:

We will then work on a project to build an AI system for a social good problem. For inspiration, here are the UN sustainable development goals for 2030.  

Fall 2021 Final Project: Food Deserts & Allocation

A food desert is an area that has limited access to affordable and nutritious food. The scope of this project is to build a delivery and distribution system to help the people in food deserts meet their nutritional needs and have a better living standard. This system that would provide a Fair, Opportunistic Resource Allocation (FORAll) to those that need it. The FORAll system learns the supply, demand, and preferences of users in an area and dynamically allocates resources so that their daily wants and needs are met.

Part 1: Decision Making

This group used Markov Decision Processes to fairly distribute food while accounting for user preferences even when demand far exceeds supply.

Part 2: Learning

The learning group's main purpose is to determine expected supply and demand vectors and an user preference/priority matrix based on prior knowledge and feedback from the communication group. 

Part 3: Communication I

This team used a Factorized Neighborhood Model with both item-based and user-based filtering to create a preference learning module for the FORAll system.

Part 4: Communication II

The Communication II group sought to predict a user's feedback about quantity of an item allocated to them