Qianli XU personal page

• Position

Scientist, A*STAR, Institute for Infocomm Research (Visual Intelligence Dept)

Computer Vision, Human-computer interaction, Reinforcement learning.

• Research topics with IPAL

Current or Intended projects:

Project 1 : Learning procedural knowledge through interaction for assisted living (AI)

This project aims to develop technologies that endow robots with the ability to learn procedural knowledge through human-in-the-loop (interactive) learning. The learning process can be designed as an iterative cycle of robot learning (establishment of mental model) and communicating the learning outcome to a human teacher via novel information display techniques. We want to train the robot not only to capture the task knowledge, but more importantly, to let the human know how much has been learned (or not learned). The later requires a quantifiable measurement of the uncertainty level at which the robot infers about a source state and a target state (e.g., “put the cup on the shelf” may be ambiguous if two cups are visible in the scene). In this case, the robot may resolve the ambiguity by asking the human teacher questions (e.g., “Do you mean the red one?”). This leads to many interesting and intriguing research questions, such as (1) How to represent task knowledge in a format that can be communicated easily between human and robot? (2) How to capture simple commonsense knowledge, such as object attributes, affordance, which can be used to restrict the task representation? (3) How to establish an uncertainty measure of robot mental model of the task? (4) How to design an interaction protocol to allow robot to ask relevant questions to communicate its learning outcome? (5) How to capture human feedback and transfer it into new knowledge?

Project 2 : Reinforcement learning for TEM parameter tuning (AI)

Using Transmission Electron Microscope (TEM), a sample at micro or nano scale can be captured by a camera or visible through a phosphor screen. A good sample image relies on adjustment on the parameters of TEM.  Thus, TEM is usually operated by domain expert for capturing desired sample images, as it has multiple and complex parameters. Applying AI technique to assist parameters adjustment on TEM can alleviate the requirement of domain expertise and reduce much operating time. Since there is no ground truth to guide the parameter’s tuning and each operation needs user’s feedback, reinforcement learning (RL) is a promising AI technique to tackle the problem of learning the parameter’s tuning model.

• Collaborators at IPAL

Potential collaborator: Julien Briand (Project 1)

Potential collaborator: Loïc Grossetête (Project 2)

• Supervisions at IPAL

TBD

• Link to webpage

You can know more about me here: https://a-star.academia.edu/QianliXu