SBIR-STTR Award

TARGET: Transfer via Active Requests to Generalize Effective Training
Award last edited on: 6/4/2021

Sponsored Program
STTR
Awarding Agency
DOD : Navy
Total Award Amount
$1,940,130
Award Phase
2
Solicitation Topic Code
N15A-T013
Principal Investigator
Robert Wray

Company Information

Soar Technology Inc (AKA: SoarTech)

3600 Green Court Suite 600
Ann Arbor, MI 48105
   (734) 627-8072
   info@soartech.com
   www.soartech.com

Research Institution

University of California - Santa Barbara

Phase I

Contract Number: N00014-15-P-1184
Start Date: 7/6/2015    Completed: 11/6/2016
Phase I year
2015
Phase I Amount
$149,884
We propose an intelligent assessment concept that will discover and reason about KSAs in multiple instruc-tional domains and settings. SoarTech will team with world-class researchers from the University of Califor-nia, Davis and the University of Memphis to research and develop TARGET: Transfer via Active Requests to Generalize Effective Training. TARGET will enhance SoarTechs assessment system DARTS, currently in schoolhouse use, with Active Transfer Learning that models student performance and skills in multiple in-structional domains, identifies underlying skills, finds relationships between domains, and actively queries human experts to continually improve its model. Learning Space Theory makes detailed cross-domain models tractable, enables inference that is understandable to nontechnical users and mathematically sound, and di-rects adaptive testing or intelligent tutoring. With these components, TARGET can predict what KSAs might be important to train a new capability, reduce testing needed to estimate students underlying proficiency, and remain accessible to instructors and SMEs who will be able to understand, trust, control, and update the sys-tem

Benefit:
TARGET will improve personnel recruitment, selection, accession, retention and professional development because it will increase the precision, accuracy, and amount of information available about individual skill proficiencies. TARGET will both increase training effectiveness and also reduce required time and costs by eliminating duplicative training that merely focuses on superficial test scores or checklists rather than actual, underlying proficiency. Finally, TARGET will enable training to be tailored to the individual and team any-where, anytime by offering a detailed and up-to-date understanding of individual and team competencies that evolves to track the changing nature of military operations.

Keywords:
Active Transfer Learning, Active Transfer Learning, Knowledge Spaces, Intelligent Tutoring, Training, Learning Spaces, Crowd sourcing

Phase II

Contract Number: N68335-17-C-0042
Start Date: 3/9/2017    Completed: 12/9/2019
Phase II year
2017
(last award dollars: 2021)
Phase II Amount
$1,790,246

Active Transfer Learning (ATL) is a machine learning approach that produces excellent accuracy and predictive power while requiring much less input data than competing approaches. SoarTech, partnered with University of California Davis, has used ATL to improve the understanding of skills and skill relationships in the Navys tactics and decision-making assessment system, DARTS. SoarTech showed in Phase I that ATL let DARTS accurately estimate mastery of sixteen different skills after inputs of only five student data points. We implemented a working prototype and evaluated it with a series of historical and simulated datasets. During Phase II, SoarTech and partners will extend the Phase I research by enhancing the integration of ATL with the DARTS system, addressing limitations in the state of the art specific to understanding complex operational skills, and using ATL to enable new capabilities in training and personnel management such as crowdsourcing to capture new knowledge and updates from operational users. We will evaluate and validate the usefulness of the ATL approach in a series of studies using simulated students and human users in the Option period. The research will lead to new, more detailed, and more frequently updated understanding of skills for training and personnel management experts.

Benefit:
The proposed research will contribute to the Navys understanding of skills Sailors need for complex tasks in training and throughout an operational career. The research will improve training by enabling Ready, Relevant Learning. TARGET will enable increased modularization of learning content by analyzing how existing and future learning modules relate to explicit and underlying skills, and enable increased tailoring of the learning trajectory by accurately estimating learners progress and needs relevant to those skills. The research will also improve operational job performance by enabling Personnel System Modernization. TARGET will monitor job performance in relation to underlying skills that cut across job tasks, and identify whose underlying skills will make them most likely to retrain when personnel are needed to fill a gap. Finally, commercialization of the TARGET research via an existing commercial intelligent tutoring system will benefit the Navy in addition to the direct benefits of military transition. This is because TARGET in the commercial system will explicitly improve teaching and training for STEM and STEAM topics in the K-12 age range. Improved student knowledge and skill in these fields will tend to improve the students readiness to contribute to the Navy as new recruits and to learn new Navy skills throughout their career.

Keywords:
Skill Assessment, Sailor 2025, Intelligent Tutoring System, learning space theory, Transfer Learning, Personnel System Modernization, active learning, Ready Relevant Learning