Aaron Mininger

Computer Scientist and Educator


I am a PhD candidate at the University of Michigan doing research in artificial intelligence by improving methods for robots to learn new tasks through natural language instruction. I am passionate about teaching, and plan on seeking a career in post-secondary education with a focus on fostering strong mentoring relationships with students and creating an active classroom environment that stresses problem-solving and practical experience.


Education

2013
Master's Degree in Computer Science
University of Michigan, MI
2011
Bachelor's Degree in Computer Science
Grove City College, PA
2010
Research Experience for Undergraduates
Harvey Mudd College, CA
2006
PA Governor's School for the Sciences
Carnegie Mellon University, PA

Research

What if we could teach robots using language?

This research question is being tackled by the field of Interactive Task Learning. My focus is on creating robotic agents that can learn new tasks in one-shot from natural language instruction. My research involves a cognitive architecture approach, where we investigate issues of knowledge representation, reasoning, and explanation-based learning.


Rosie

Rosie is the name of the Interactive Task Learning agent I have helped develop at the University of Michigan in the Soar Lab. It is written in the Soar Cognitive Architecture. It can learn many different tasks, games, and puzzles and has been deployed in many different real and simulated environments. Here is a demo video of it learning tasks in a mobile environment.

Domains

Tabletop Arm
Magicbot
Simulator
Fetch Robot

Publications

Open Pdf 2019 | Aaron Mininger, John E. Laird Mininger, Laird
Demonstrates how an agent in a symbolic cognitive architecture with access to a spatial short-term memory can use spatial reasoning and domain knowledge to participate in the anchoring process and to detect and correct bottom-up anchoring errors in its input.
Open Pdf 2019 | John E. Laird, Shiwali Mohan, James Kirk, Aaron Mininger Laird, Mohan, Kirk, Mininger
Identifies and describes the key characteristics of the learning problem inherent to Situated Interactive Task Learning.
Open Pdf 2018 | Aaron Mininger, John E. Laird Mininger, Laird
Explains how the task-learning part of Rosie was extended to also learn simple procedural tasks in an integrated manner.
Open Pdf 2017 | Peter Lindes, Aaron Mininger, James Kirk, John E. Laird Lindes, Mininger, Kirk, Laird
Describes a new language comprehension system called Lucia based on Embodied Construction Grammar that is integrated with Rosie.
Open Pdf 2016 | Aaron Mininger, John E. Laird Mininger, Laird
Describes work done to extend Rosie to a mobile environment and how it handles references to objects that are not visible. Rosie learns different strategies for finding objects through instruction.
Open Pdf 2016 | James Kirk, Aaron Mininger, John E. Laird Kirk, Mininger, Laird
Describes work that allows Rosie to learn task goals through visual demonstrations instead of linguistic descriptions. This was done with games, puzzles, and regular tasks on the tabletop robot.
Open Pdf 2015 | Shiwali Mohan, James Kirk, Aaron Mininger, John E. Laird Mohan, Kirk, Mininger, Laird
Identifies and describes nine system-level requirements for collaborative robotic agents that help them support more effective, efficient, and task-oriented dialog.
Open Pdf 2014 | Shiwali Mohan, Aaron Mininger, John E. Laird Mohan, Mininger, Laird
Describes a computational model of situated language comprehension that incorporates knowledge from perception, domain knowledge, and short and long term experiences when generating semantic representations.
Open Pdf 2012 | Shiwali Mohan, Aaron Mininger, James Kirk, John E. Laird Mohan, Mininger, Kirk, Laird
Describes an early version of the Rosie project with the tabletop blocks-world domain that can learn nouns (shapes), adjectives (colors), prepositions (spatial relations), and verbs (goal-based actions).
Open Pdf 2012 | Shiwali Mohan, Aaron Mininger, James Kirk, John E. Laird Mohan, Mininger, Kirk, Laird
Describes an early version of the Rosie project with the tabletop blocks-world domain that can learn nouns (shapes), adjectives (colors), prepositions (spatial relations), and verbs (goal-based actions).

Projects

Below is a list of some of the projects that I have worked on in my spare time over the last several years. Many of these are hosted on my Github Account.

A lego robot project used as the tutorial for the 2014 Soar Workshop. Participants could program a soar agent that controlled the robot over wifi to solve simple mazes.
A simple Chrome extension that creates a list view of upcoming events and provides a text interface for adding new events to Google Calendar.
A python library that makes it easier to create soar agents within python and interface with them using SML.
A vim plugin that adds Soar editing features and provides a way to run and debug soar agents inside of vim.
An online recipe database I created featuring favorite recipes that I've collected over time.

Teaching

I strive to be a computer science professor who is a
  • Comprehensive educator that develops well rounded computer science professionals with skills in problem-solving, communication, and teamwork.
  • Relational instructor that creates a supportive classroom environment and personally invests in students.
  • Lifelong learner that refines teaching methods in response to observations, feedback, and educational research.

Prior Experience

2017
Lecturer for EECS 280
Programming and Introductory Data Structures
2017
Preparing Future Faculty Seminar
Center for Research on Learning and Teaching
2016
Graduate Student Instructor for EECS 592
Advanced Artificial Intelligence