Inferring Personalized Bayesian Embeddings for Learning from Heterogeneous Demonstration
This addresses the challenge of capturing individual differences in human behavior for assistive robots and virtual agents, representing an incremental improvement over traditional LfD methods.
The paper tackled the problem of human heterogeneity in decision-making for Learning from Demonstration (LfD) by proposing a Bayesian framework that infers personalized embeddings, and it validated this approach by outperforming state-of-the-art techniques on synthetic and real-world datasets.
For assistive robots and virtual agents to achieve ubiquity, machines will need to anticipate the needs of their human counterparts. The field of Learning from Demonstration (LfD) has sought to enable machines to infer predictive models of human behavior for autonomous robot control. However, humans exhibit heterogeneity in decision-making, which traditional LfD approaches fail to capture. To overcome this challenge, we propose a Bayesian LfD framework to infer an integrated representation of all human task demonstrators by inferring human-specific embeddings, thereby distilling their unique characteristics. We validate our approach is able to outperform state-of-the-art techniques on both synthetic and real-world data sets.