Uncertainty estimation improves proactive assistant user preference alignment
Vision: "What Don't I Know About User's Preferences?" How Uncertainty Estimation Improves Proactive Assistant Alignment
Artificial IntelligenceHuman-Computer Interaction
Summary
Proactive smart assistants try to guess what users want before they ask. However, simply predicting preferences doesn't tell the assistant how confident it is or why some predictions are unclear. The authors suggest that measuring uncertainty can help assistants understand the confidence and sources of their guesses, leading to better decisions. They present examples from smart homes and look at related areas like activity recognition and human movement prediction to show how uncertainty awareness can improve assistants.
What this means in practice
- •For smart home developers: Design home assistants that assess prediction confidence to better handle ambiguous user preferences and adapt actions accordingly.
- •For wearable device engineers: Incorporate uncertainty metrics in activity recognition systems to improve personalization and reliability based on available user data.
A position paper. It proposes an approach and reports no results.
Authors
Zhiyu An, Ziyi Xuan, Yu Yang, Wan Du
Abstract
Proactive Smart Assistants (PSAs) increasingly use general knowledge, learned patterns from group-level data, patterns from individual user's data, and recent sensor observations to anticipate which action a user may prefer. Existing systems typically select actions from predicted user preference likelihoods, but these likelihoods alone do not reveal how reliable the prediction is, which evidence supports it, or why ambiguity remains. In this vision paper, we argue that uncertainty estimation can provide this missing information and substantially change appropriate downstream behavior. Using the example of a home assistant, we introduce the questions that uncertainty estimation aims to answer, i.e. how sure is the assistant about the predicted preference likelihood, whether the confidence come from group-level assumptions or individual's data, and whether likelihood ambiguity reflect the user's intrinsic preference variability or insufficient data for the particular situation. We further introduce uncertainty estimation applications in neighboring domains, such as human trajectory prediction, personalized activity recognition, information-collection, and model-based control. Finally, we draw on insights from neighboring domains to envision how future uncertainty-aware proactive assistants may address key challenges in proactive interaction.