Optimal set recommendations based on regret
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Current conversational recommender systems do not offer guarantees on the quality of their recommendations, either because they do not maintain a model of a user's utility function, or do so in an ad hoc fashion. In this paper, we propose an approach to recommender systems that incorporates explicit utility models into the recommendation process in a decision-theoretically sound fashion. The system maintains explicit constraints on the user's utility based on the semantics of the preferences revealed by the user's actions. In particular, we propose and investigate a new decision criterion, setwise maximum regret, for constructing optimal recommendation sets. This new criterion extends the mathematical notion of maximum regret used in decision theory and preference elicitation to sets. We develop computational procedures for computing setwise max regret. We also show that the criterion suggests choice sets for queries that are myopically optimal: that is, it refines knowledge of a user's utility function in a way that reduces max regret more quickly than any other choice set. Thus setwise max regret acts both as guarantee on the quality of our recommendations and as a driver for further utility elicitation. Our simulation results suggest that this utility-theoretically sound approach to user modeling allows much more effective navigation of a product space than traditional approaches based on, for example, heuristic utility models and product similarity measures.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it