Lin Guan
lguan9 [at] asu [dot] edu

I'm a 4th-year Ph.D. student in Computer Science at Arizona State University. I work at Yochan Lab (AI Lab), supervised by Dr. Subbarao Kambhampati. My research interests lie at the intersection of machine learning (especially reinforcement learning from human feedback and LLMs for task planning) and human-AI interaction. Specifically, I am working on:

(a) Developing algorithms that enable human users to effortlessly specify desired AI-agent behaviors through human inputs such as preference labels (i.e., learning from human feedback or RLHF), symbolic or natural language feedback, etc. Applications include aligning AI-model behaviors with human preferences, character/animation control, autonomous vehicles, and robot skill learning.

(b) Building intelligent agents that can learn to solve long-horizon tasks by leveraging pre-trained large language models (LLMs) for task planning, or by integrating planning with reinforcement learning.

email   |   cv   |   google scholar   |   twitter   |   linkedIn   |   github

• [2023.01]

  1 paper accepted to ICLR 2023.

• [2022.05]

  Student researcher at Google.

• [2022.04]

  1 paper accepted to ICML 2022 (also accepted to RLDM 2022 and received the Best Paper Award at PRL@ICAPS 2022).

• [2021.10]

  1 paper accepted to NeurIPS 2021 as a Spotlight presentation (3%).

• [2021.05]

  Machine learning software engineer intern at TikTok and worked on AutoML.

[Older News]

[2021.10]

1 paper accepted to AAAI 2022 Blue Sky Track.

[2021.06]

1 paper accepted to IROS 2021.

[2019.12]

1 paper accepted to AAAI 2020.

[2019.04]

1 paper accepted to IJCAI 2019.

To leverage LLMs in planning tasks, we introduce an alternative paradigm that teases an explicit world (domain) model in PDDL out of LLMs and then uses it to plan with sound domain-independent planners. This is motivated by the insight that LLMs, while incapable of the combinatorial search needed to produce correct plans, may be better suited as the source of world models.

We introduce the notion of Relative Behavioral Attributes that enables end users to tweak the agent's behavior through nameable concepts even for a tacit-knowledge skill learning task (e.g., decreasing the steering sharpness of an autonomous driving agent, or increasing the softness of the movement of a two-legged "sneaky" agent).

Symbolic knowledge is important for solving long-horizon task and motion planning tasks. But a key resistance to leveraging easily available human symbolic knowledge has been that it might be inexact. In this work, we present a framework to quantify the relationship between the true task model and an inexact STRIPS model, and introduce a novel approach using landmarks and a diversity objective to make up for potential errors in the symbolic knowledge.

We make Human-in-the-Loop RL more efficient and feasible by allowing human teachers to not only give binary feedback but also to highlight task-relevant features.

We advocated an ambitious research program to broaden the basis for human-AI interaction by proposing that AI systems support a symbolic interface – independent of whether their internal operations themselves are done in human-interpretable symbolic means.

We provide a large-scale, high-quality dataset of human actions with simultaneously recorded eye movements (i.e., gaze info) while humans play Atari video games.

This survey provides a high-level overview of five learning frameworks that primarily rely on human guidance other than conventional, step-by-step action demonstrations.

• [2023]

  Served as a reviewer for ICML 2023 and NeurIPS 2023.

• [2022]

  Served as a reviewer for NeurIPS 2022.

• [2021]

  Served as program committee for AAAI 2022 and ICRA 2022.

• [2020]

  Served as program committee for AAAI 2021.