Seminars and Events
Building Generalizable, Efficient, and Trustworthy Multimodal Embodied Agents
Event Details
A long-term goal of AI research is to build intelligent agents that can communicate with humans, perceive multimodal environments, and act in the real world. Towards building such agents, this talk will cover our recent work in three key areas: multimodal representation learning, embodied agents for human-agent communication, and ethical considerations of intelligent agents. First, multimodal representation learning is a critical foundation for building intelligent agents, as it allows agents to perceive and understand the multimodal world through multiple sources of information, such as visual, auditory, and linguistic data. Second, we are particularly interested in developing generalized embodied agents for human-agent communication. These agents must be able to learn and reason about their surroundings, as well as communicate and interact with humans naturally and intuitively, to perform real-world multimodal tasks. Furthermore, trustworthiness, fairness, and privacy are essential considerations of intelligent agents. As these agents will be interacting with humans closely, it is crucial that they are designed to operate in an ethical and transparent manner, and are trustworthy and beneficial to society.
Speaker Bio
Xin (Eric) Wang is an Assistant Professor of Computer Science and Engineering at UC Santa Cruz. His research interests include Natural Language Processing, Computer Vision, and Machine Learning, with a focus on building embodied AI agents that can communicate with humans through natural language to perform real-world multimodal tasks. Xin has served as an Area Chair for conferences such as ACL, NAACL, EMNLP, and ICLR, as well as a Senior Program Committee for AAAI and IJCAI. I have also organized workshops and tutorials at conferences such as ACL, NAACL, CVPR, and ICCV. He has received several awards and recognitions for his work, including the CVPR Best Student Paper Award (2019) and a Google Research Faculty Award (2022), as well as three Amazon Alexa Prize Awards (2022-2023).
YOU ONLY NEED TO REGISTER ONCE TO ATTEND THE ENTIRE SERIES – We will send you email announcements with details of the upcoming speakers.
Register in advance for this webinar: https://usc.zoom.us/webinar/register/WN__0VhakI6Q6i3JsasdmNWcA
After registering, you will receive an email confirmation containing information about joining the Zoom webinar.
The speaker approved to be recorded for this AI Seminar talk; it will ONLY be available internally using your USC or ISI domain via the link below https://usc.zoom.us/rec/share/jrjIZP-UZYuzLr2Ir-tKhEdOWLjHLHyWc9N0n66KbVpk9dIJzEh3qnIyJIlnM9_z.L9xHLHIIpYbjZZKx?startTime=1677265339000
Host: Mohammad Rostami, POC: Mable Johnson