Related Experiment Video
Updated: Nov 1, 2025
11:54
Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface
Published on: May 8, 2021
4.8K
Lifelong Personalization via Gaussian Process Modeling for Long-Term HRI
Samuel Spaulding1, Jocelyn Shen1, Hae Won Park1
1Massachusetts Institute of Technology, Cambridge, MA, United States.
Frontiers in Robotics and AI
|June 24, 2021
Summary
Lifelong Personalization combines continual and multitask learning for adaptive AI agents. This approach improves personalization by actively managing training data in dynamic, long-term user interactions.
Area of Science:
- Artificial Intelligence
- Human-Computer Interaction
- Machine Learning
Background:
- Artificial agents that adapt to users can transform social services.
- Long-term user interactions require advanced personalization models.
- Continual learning and multitask personalization are key research areas.
Purpose of the Study:
- To unite continual learning and multitask personalization under the framework of Lifelong Personalization.
- To model dynamic learners over time in interactive teaching scenarios.
- To develop a method for active and continual management of training data.
Main Methods:
- Augmenting a Gaussian Process-based multitask personalization model.
- Implementing a mechanism for active training data management (removal/weight reduction).
- Evaluating the method using simulation experiments with dynamic student data.
Main Results:
- The proposed Lifelong Personalization framework effectively models dynamic learners.
- Active training data management improves learning in dynamic domains.
- Gaussian Processes show promise as a flexible tool for long-term HRI.
Conclusions:
- Lifelong Personalization offers a robust approach for adaptive AI in long-term interactions.
- Active data management is crucial for personalization in non-stationary environments.
- This work extends the applicability of Gaussian Processes in Human-Robot Interaction.
Related Concept Videos
01:28Automatic Processing and Automatic Social Behavior
76
Automatic processing refers to the cognitive operations that occur without conscious intent or awareness, playing a fundamental role in shaping social cognition and behavior. These processes enable individuals to navigate complex social environments efficiently by relying on mental shortcuts and pre-existing knowledge structures known as schemas. One of the most influential mechanisms underlying automatic processing is priming, which subtly activates mental representations through exposure to...
76
02:16Stereotype Content Model
15.0K
The Stereotype Content Model (SCM) was first proposed by Susan Fiske and her colleagues (Fiske, Cuddy, Glick & Xu, 2002; see also Fiske, 2012 and Fiske, 2017). The SCM specifies that when someone encounters a new group, they will stereotype them based on two metrics: warmth—or that group’s perceived intent, and how likely they are to provide help or inflict harm—and competence—or their ability to carry out that objective. Depending on the warmth-competence...
15.0K