MTMichael Tamir

Michael Tamir

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Michael Tamir has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    TubiFM: Unified Item, Carousel, and Search Ranking for Streaming Discovery

    by Alexandre Salle, Chenglei Niu, Suchismit Mahapatra, Xiaoxiao Chen, Suvash Sedhain, Yaqi Wang, Shervin Shahryari, Saurabh Agrawal, Qiang Chen, Michael Tamir

    Personalized discovery systems often train separate models for item ranking, carousel ranking, and search, even though these tasks expose complementary signals from the same viewer journey: watches shape carousel and item ranking, search queries reveal intent even when they do not lead to a catalog match, and watch history helps interpret search as rewatching, continuation, or new discovery. We introduce the user story, a serialized representation that turns a user’s cross-surface history (attributes, sessions, watch events with surface and carousel context, and search events) into a single token sequence. By interleaving pretrained language tokens with domain-specific event tokens, user stories let heterogeneous recommendation and search tasks be expressed as prompted next-token prediction over a shared grammar. TubiFM is one instantiation of this approach: a Llama 3.2 1B-based model trained on user stories and prompted to rank items, carousels, or search results without task-specific architectures. In offline evaluation, this single model outperforms specialist baselines across item, carousel, and search ranking. In online A/B tests, TubiFM significantly improves search total viewing time (TVT) by +3.9% and carousel TVT by +0.30%. Item ranking is statistically neutral on TVT (+0.14%), matching a mature production stack while replacing it with a single model. Across all three tasks, TubiFM serves on L40S GPUs and reduces p99 ranking latency from approximately 500ms to 200ms. These results show that shared user stories can improve discovery while simplifying ranking systems.

  • Open Access

    The Curve Fitting Problem, Data Validation, and Inductive Generalization in Machine Learning

    by Michael Tamir, Elay Shech

    Aris Spanos and Deborah Mayo’s error-statistical approach to statistical modeling and inference adopts the reliability of inductive inference as a primary criterion for statistical model and estimator selection (e.g., curve fitting). In this paper, we expand the error-statistical approach’s adoption of reliable inductive inference by scrutinizing the epistemic legitimacy of contemporary techniques leveraged in data science. We argue that data validation and testing potentially provides a direct, measurable method of evaluating evidence for reliable inductive inferences in cases where the error-statistical approach is not easily applied, and conclude with an exploration of core methodological foils to the reliability of inductive inference revealed by this argument.


  • Open Access

    Machine understanding and deep learning representation

    by Michael Tamir, Elay Shech

    Practical ability manifested through robust and reliable task performance, as well as information relevance and well-structured representation, are key factors indicative of understanding in the philosophical literature. We explore these factors in the context of deep learning, identifying prominent patterns in how the results of these algorithms represent information. While the estimation applications of modern neural networks do not qualify as the mental activity of persons, we argue that coupling analyses from philosophical accounts with the empirical and theoretical basis for identifying these factors in deep learning representations provides a framework for discussing and critically evaluating potential machine understanding given the continually improving task performance enabled by such algorithms.


  • Understanding from Deep Learning Models in Context

    by Michael Tamir, Elay Shech

    This chapter places into context how the term model in machine learning (ML) contrasts with traditional usages of scientific models for understanding and we show how the direct analysis of an estimator’s learned transformations (specifically, the hidden layers of a deep learning model) can improve understanding of the target phenomenon and reveal how the model organizes relevant information. Specifically, three modes of understanding will be identified: the difference between implementation irrelevance and functionally approximate irrelevance will be disambiguated, and how this distinction impacts potential understanding with these models will be explored. In addition, by distinguishing between empirical link failures and representational ones, an ambiguity in the concept of link uncertainty will be addressed, thus clarifying the role played by scientific background knowledge in enabling understanding with ML.


  • Expecting Too Much from Our Machine Learning Models

    by Elay Shech, Michael Tamir

    This chapter responds to Emily Sullivan’s chapter “How Values Shape the Machine Learning Opacity Problem” in this volume.