What are you optimizing for? Aligning Recommender Systems with Human Values

Abstract

We describe cases where real recommender systems were modified in the service of various human values such as diversity, fairness, well-being, time well spent, and factual accuracy. From this we identify the current practice of values engineering: the creation of classifiers from human-created data with value-based labels. This has worked in practice for a variety of issues, but problems are addressed one at a time, and users and other stakeholders have seldom been involved. Instead, we look to AI alignment work for approaches that could learn complex values directly from stakeholders, and identify four major directions: useful measures of alignment, participatory design and operation, interactive value learning, and informed deliberative judgments.

arXiv commentOriginally presented at the ICML 2020 Participatory Approaches to Machine Learning workshop

References in corpus (3)

• RecoGym: A Reinforcement Learning Environment for the problem of Product Recommendation in Online Advertising
• Safe Imitation Learning via Fast Bayesian Reward Inference from Preferences
• Active Preference-Based Gaussian Process Regression for Reward Learning

Cited by in corpus (6)

• Unsolved Problems in ML Safety
• Recommender Systems for Sustainability: Overview and Research Issues
• Respect for Human Autonomy in Recommender Systems
• The Value-Sensitive Conversational Agent Co-Design Framework
• Beyond Algorethics: Addressing the Ethical and Anthropological Challenges of AI Recommender Systems
• Tournesol: A quest for a large, secure and trustworthy database of reliable human judgments