Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A4L: An Architecture for AI-Augmented Learning

Ashok Goel, Ploy Thajchayapong, Vrinda Nandan, Harshvardhan Sikka, Spencer Rugaber

TL;DR

The paper addresses the need for dedicated data architectures to enable AI-augmented learning at scale for adult online education. It proposes the Architecture for AI-Augmented Learning (A4L), a data pipeline and model that ingests, anonymizes, transforms, stores, analyzes, and visualizes diverse learning data to continuously personalize instruction. Preliminary applications with VERA, Jill Watson, and SAMI illustrate micro- and meso-learning, along with bi-directional feedback loops between instructors, learners, and AI agents. The work anticipates A4L2.0 to broaden data ingestion and standardization toward macro-learning and cross-institutional data sharing while preserving privacy, with NSF support guiding future development and impact on scalable personalized education.

Abstract

AI promises personalized learning and scalable education. As AI agents increasingly permeate education in support of teaching and learning, there is a critical and urgent need for data architectures for collecting and analyzing data on learning, and feeding the results back to teachers, learners, and the AI agents for personalization of learning at scale. At the National AI Institute for Adult Learning and Online Education, we are developing an Architecture for AI-Augmented Learning (A4L) for supporting adult learning through online education. We present the motivations, goals, requirements of the A4L architecture. We describe preliminary applications of A4L and discuss how it advances the goals of making learning more personalized and scalable.

A4L: An Architecture for AI-Augmented Learning

TL;DR

The paper addresses the need for dedicated data architectures to enable AI-augmented learning at scale for adult online education. It proposes the Architecture for AI-Augmented Learning (A4L), a data pipeline and model that ingests, anonymizes, transforms, stores, analyzes, and visualizes diverse learning data to continuously personalize instruction. Preliminary applications with VERA, Jill Watson, and SAMI illustrate micro- and meso-learning, along with bi-directional feedback loops between instructors, learners, and AI agents. The work anticipates A4L2.0 to broaden data ingestion and standardization toward macro-learning and cross-institutional data sharing while preserving privacy, with NSF support guiding future development and impact on scalable personalized education.

Abstract

AI promises personalized learning and scalable education. As AI agents increasingly permeate education in support of teaching and learning, there is a critical and urgent need for data architectures for collecting and analyzing data on learning, and feeding the results back to teachers, learners, and the AI agents for personalization of learning at scale. At the National AI Institute for Adult Learning and Online Education, we are developing an Architecture for AI-Augmented Learning (A4L) for supporting adult learning through online education. We present the motivations, goals, requirements of the A4L architecture. We describe preliminary applications of A4L and discuss how it advances the goals of making learning more personalized and scalable.
Paper Structure (32 sections, 7 figures)

This paper contains 32 sections, 7 figures.

Table of Contents

  1. Introduction
  2. Ecosystems for Adult Learning and Online Education
  3. An Ecosystem for Online Education
  4. An Ecosystem for Adult Learning
  5. Data Requirements for Online Education for Adult Learners
  6. Micro-learning (Individual Learning Episodes):
  7. Meso-learning (Learning Over Time):
  8. Macro-learning (Long-Term Educational Progress):
  9. The Architecture for AI-Augmented Learning (A4L)
  10. Data Sources
  11. Data Engine
  12. Data Upload and Standardization
  13. Data Anonymization
  14. Data Transformation
  15. Data Store
  16. ...and 17 more sections

Figures (7)

  • Figure 1: Elements of the AI-ALOE ecosystem for online learning and teaching. (The dashed arrows on the top left in figure indicate that the relationship is optional.)
  • Figure 2: Translating Micro-, Meso-, and Macro-Learning into data requirements.
  • Figure 3: The A4L Data Pipeline Conceptual Architecture.
  • Figure 4: Conceptual Data Model for linking student data across AI tools. (The AI agents in green have been deployed at Georgia Tech (GT) while those in yellow have been deployed at the Technical College System of Georgia (TCSG).)
  • Figure 5: Extensions to the A4L Data Pipeline include Analytics and Visualization modules on the right
  • ...and 2 more figures