Michael S. Lee

Hi, I am a post-doc in the Robotics Institute at Carnegie Mellon University, working at the intersection of explainable AI and human-AI interaction.
I am fortunate to be supported by Reid Simmons and Henny Admoni, and am part of the Human and Robot Partners (HARP) Lab.

As AI agents increasingly enter human society, it is paramount that their decision making and behaviors are transparent (i.e. understandable and predictable) to humans. Thus, I research how AI agents can select and show its key decisions and behaviors to humans, toward helping humans generalize to the AI's overall decision making in a particular domain.

I previously completed a Master's at Carnegie Mellon in robotics with Red Whittaker and Nathan Michael researching autonomous radiation source localization. And before that, I graduated from Princeton University ('16) with degrees in Mechanical & Aerospace Engineering and a certificate in Robotics & Intelligent Systems. My undergraduate thesis in computer vision on modeling uncertainty in feature-based visual odometry was advised by Nathan Michael and Jianxiong Xiao.

I plan to transition into industry in 2024, and am currently looking for opportunities related to my research interests (including but not limited to explainable AI, AI alignment/safety, human-AI interaction, and robotics).

Email  /  Google Scholar  /  LinkedIn

profile photo
Research

Impact: Throughout my PhD, I published at top AI, machine learning, and robotics conferences (e.g. AAAI, IROS, HRI), workshops (e.g. ICML), and journals (Frontiers in Robotics and AI, THRI), and was selected into two doctoral consortiums (AAAI, HRI). I have also had the fortune of of collaborating with researchers at multiple institutions throughout my PhD (Google Deepmind, UMass Lowell, Tufts, BYU) which led to two joint publications.

Toward greater transparency (i.e. understandability and predictability) AI behaviors to humans, my research explores how an AI agent may teach its underlying reward function to a human learner using informative demonstrations (i.e. ones that exhibit key tradeoffs in decision making).

Our first key insight is that a demonstration's informativeness to a human is not intrinsic, but is inherently tied to that human's prior beliefs and their current expectations of agent behavior. We thus rely on inverse reinforcement learning and counterfactual reasoning (i.e. the extent to which the agent's demonstration differs from the human's current expectations) to evaluate a candidate demonstration's informativeness to a human at revealing the agent's reward function.

Our second key insight is that informativeness and difficulty of comprehension are often correlated for humans, and we leverage ideas from the education literature (e.g. zone of proximal development / "Goldilocks" principle) to ensure that the selected demonstrations present the right level of challenge. Too small of a difference and the reconciliation in the human's mind is trivial, and too large of a difference and the gap is irreconcilable in one shot; we thus use scaffolding to provide demonstrations that incrementally increase in information gain and simultaneously ease humans into learning.

Finally, we explore how to select a suite of informative tests (which query the human's ability to correctly predict agent behavior unseen scenarios) that reveal and bridge remaining gaps in the human learner's understanding, which can then be further supported through subsequent targeted demonstrations in a closed-loop fashion.

Improving the Transparency of Agent Decision Making Using Demonstrations
Michael S. Lee,
PhD Thesis, 2024
pdf / code / slides

Developing algorithms for teaching AI policies to humans using informative demonstrations of AI behavior. Conducted four user studies involving 750+ participants. Our teaching model reduces the suboptimality of human predictions of AI behavior by 64% over the baseline of directly providing the AI’s reward function.

Closed-loop Teaching via Demonstrations to Improve Policy Transparency
Michael S. Lee, Reid Simmons, Henny Admoni
Pre-print
pdf / supplemental

Designing a closed-loop teaching framework where AI policy is made more transparent to a human via demonstrations of AI behavior, tests, and feedback. A novel particle filter model of human beliefs is maintained to provide demonstrations that are targeted to the human's current understanding in real time.

Making AI Policies Transparent to Humans through Demonstrations
Michael S. Lee
AAAI Doctoral Consortium, 2024
pdf

A 2-page research statement summarizing my dissertation research to date and highlighting potential directions for future work.

Closed-loop Reasoning about Counterfactuals to Improve Policy Transparency
Michael S. Lee, Henny Admoni, Reid Simmons
ICML Workshop on Counterfactuals in Minds and Machines, 2023
pdf / poster

Reasoning over a human's counterfactual expectations of the AI's policy in real time to provide informative demonstrations of AI behavior that differ meaningfully.

Leveraging Contextual Counterfactuals Toward Belief Calibration
Qiuyi (Richard) Zhang, Michael S. Lee, Sherol Chen
ICML Workshop on Counterfactuals in Minds and Machines, 2023
pdf / poster

Leveraging counterfactual reasoning over possible outcomes and recourses to identify optimal belief strengths (i.e. parameters) for algorithmic decision making.

Reasoning about Counterfactuals to Improve Human Inverse Reinforcement Learning
Michael S. Lee, Henny Admoni, Reid Simmons
International Conference on Intelligent Robots and Systems (IROS), 2022
pdf / code / user study

An informative demonstration is one that differs meaningfully from the learner’s expectations of what the robot will do given their current understanding of the robot’s decision making.

Building the Foundation of Robot Explanation Generation Using Behavior Trees
Zhao Han, Daniel Giger, Jordan Allspaw, Henny Admoni, Michael S. Lee, Holly Yanco
ACM Transactions on Human-Robot Interaction (THRI), 2022
pdf / code / project page / slides

A behavior tree hierarchically composed of goals, subgoals, steps, and actions supports explanation generation algorithms that convey causal information about robot behavior.

Machine Teaching for Human Inverse Reinforcement Learning
Michael S. Lee, Henny Admoni, Reid Simmons
Frontiers in Robotics and AI, 2021
pdf / code / user study

Augmenting an inverse reinforcement learning model of how humans learn from demonstrations with teaching strategies (e.g. scaffolding, simplicity, pattern discovery, and testing) better accommodates human learners.

Self-Assessing and Communicating Manipulation Proficiency Through Active Uncertainty Characterization
Michael S. Lee,
Pioneers Workshop at ACM/IEEE International Conference on Human-Robot Interaction (HRI), 2019
pdf

A proposal for how a robot can actively assess both its proficiency at a task and its confidence in that assessment through appropriate measures of uncertainty that can be efficiently and effectively communicated to a human.

I previously developed a novel gamma radiation map representation, source localization algorithm, and frontier-based exploration for efficient radiological characterization of nuclear facilities using a robot equipped with a gamma-ray camera.

Radiation Source Localization using a Gamma-ray Camera
Michael S. Lee
Master's Thesis, 2018
pdf / slides

The proposed frontier-based exploration method biases frontier selection with the observed radiation field gradient to quickly search an environment until a proximal source is detected.

Active Range and Bearing-based Radiation Source Localization
Michael S. Lee, Daniel Shy, Red Whittaker, Nathan Michael
International Conference on Intelligent Robots and Systems (IROS), 2018
pdf / slides

The proposed active source localization algorithm greedily selects new waypoints that maximize the Fisher Information provided by the gamma-ray camera’s range and bearing observations.

3-D Volumetric Gamma-ray Imaging and Source Localization with a Mobile Robot
Michael S. Lee Matthew Hanczor, Jiyang Chu, Zhong He, Nathan Michael, Red Whittaker
Waste Management Conference, 2018
pdf / slides

A ground robot equipped with a Compton gamma camera localizes multiple gamma radiation sources to within an average of 0.26 m or 0.6% of the environment dimensions in two 5×4 m2 and 14×6 m2 laboratory environments.



Design by Jon Barron.