BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CREST - ECPv5.1.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:CREST
X-ORIGINAL-URL:https://crest.science
X-WR-CALDESC:Events for CREST
BEGIN:VTIMEZONE
TZID:Europe/Helsinki
BEGIN:DAYLIGHT
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:EEST
DTSTART:20260329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:EET
DTSTART:20261025T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;VALUE=DATE:20260316
DTEND;VALUE=DATE:20260331
DTSTAMP:20260709T203938
CREATED:20251210T123817Z
LAST-MODIFIED:20260324T072239Z
UID:18644-1773619200-1774915199@crest.science
SUMMARY:Taiji Suzuki (Univ. Tokyo) - Feature learning perspective of deep foundation models: statistical and optimization theories
DESCRIPTION:FEATURE LEARNING PERSPECTIVE OF DEEP FOUNDATION MODELS: STATISTICAL AND OPTIMIZATION THEORIES\nTAIJI SUZUKI\nUniversity of Tokyo\nDepartment of Mathematical Informatics \nSchedule: \nMarch 16\, 2026 | 2PM – 5PM – Room 2005\nMarch 19\, 2026 | 2PM – 5PM – Room 2005\nMarch 26\, 2026 | 2PM – 5PM – Room 2005\nMarch 30\, 2026 | 2PM – 5PM – Room 2005 \nContent:\nThe main focus of this lecture will be on the feature learning aspects of deep foundation models\, especially about the benefit of feature learning to achieve better predictive accuracy and more efficient optimization. As the deep foundation models are developed following the scaling law\, a theoretical understanding of the learning principles behind practice is getting more important. For superior generalization of deep models\, it is essentially important to acquire compressed representations avoiding mere memorization\, making representation/feature learning fundamental. It has been theoretically shown that deep learning gains various advantages in generalization via its feature learning ability which naturally arises from its deep structure. It will be discussed how the feature learning ability affects the rate of convergence by comparing it with the sub-optimal rate of non-feature learning methods. This issue will be discussed not only from statistical  perspective but also from optimization perspective. Interesting examples include estimating a Gaussian single index model in which the computational complexity can be characterized by quantities socalled information exponent and generative exponent. If time permits\, optimization guarantees by mean field Langevin dynamics and its statistical property will also be discussed. Furthermore\, feature learning is significant not only during pre-training but also during test-time inference. This will be demonstrated concisely using in-context learning as an example. It will be discussed how the test time feature learning as well as pre-training feature learning affects the performance of test time inference. In summary\, the following topics will be covered in the lecture (but some of them could be omitted depending on time constraint):\n• Nonparametric function estimation by deep learning on high dimension data and its minimax optimality.\n• Stochastic gradient descent for neural network training; Gaussian single index model\, k-parity problem\, information exponent\, CSQ/SQ lower bound.\n• Mean field Langevin dynamics and its statistical property.\n• Test time inference and test time scaling: Transformer\, in-context learning\, chain-ofthought. \nEvaluation: \nTake home project. \nReferences:\nE. Giné and R. Nickl. Mathematical foundations of infinite-dimensional statistical models. Cambridge Series in Statistical and Probabilistic Mathematics. Cambridge University Press\, 2015.\nS. Hayakawa and T. Suzuki. On the minimax optimality and superiority of deep neural network learning over sparse parameter spaces. Neural Networks\, 123:343–361\, 2020.\nJ. Kim\, T. Nakamaki\, and T. Suzuki. Transformers are minimax optimal nonparametric incontext learners. In Advances in Neural Information Processing Systems\, volume 37\, pages 106667–106713\, 2024.\nJ. D. Lee\, K. Oko\, T. Suzuki\, and D. Wu. Neural network learns low-dimensional polynomials with SGD near the information-theoretic limit. In Advances in Neural Information Processing Systems\, volume 37\, pages 58716–58756\, 2024.\nN. Nishikawa\, Y. Song\, K. Oko\, D. Wu\, and T. Suzuki. Nonlinear transformers can perform inference-time feature learning. In Proceedings of the 42nd International Conference on Machine Learning\, volume 267 of Proceedings of Machine Learning Research\, pages 46554– 46585. PMLR\, 2025.\nA. Nitanda\, D. Wu\, and T. Suzuki. Convex analysis of the mean field Langevin dynamics. In Proceedings of The 25th International Conference on Artificial Intelligence and Statistics\, volume 151 of Proceedings of Machine Learning Research\, pages 9741–9757. PMLR\, 28–30 Mar 2022.\nNitanda\, A. Lee\, D. T. X. Kai\, M. Sakaguchi\, and T. Suzuki. Propagation of chaos for mean-field Langevin dynamics and its application to model ensemble. In Forty-second International Conference on Machine Learning\, 2025.\nK. Oko\, Y. Song\, T. Suzuki\, and D. Wu. Pretrained transformer efficiently learns lowdimensional target functions in-context. In Advances in Neural Information Processing Systems\, volume 37\, pages 77316–77365\, 2024.\nT. Suzuki. Adaptivity of deep ReLU network for learning in Besov and mixed smooth Besov spaces: optimal rate and curse of dimensionality. In International Conference on Learning Representations\, 2019.\nT. Suzuki and A. Nitanda. Deep learning is adaptive to intrinsic dimensionality of model smoothness in anisotropic Besov space. In Advances in Neural Information Processing Systems\, volume 34\, pages 3609–3621\, 2021. A. B. Tsybakov. \n  \n
URL:https://crest.science/event/taiji-suzuki-univ-tokyo-feature-learning-perspective-of-deep-foundation-models-statistical-and-optimization-theories/
CATEGORIES:Doctoral Courses,Statistics
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20260325T100000
DTEND;TZID=Europe/Helsinki:20260325T110000
DTSTAMP:20260709T203938
CREATED:20260310T084040Z
LAST-MODIFIED:20260310T091904Z
UID:18855-1774432800-1774436400@crest.science
SUMMARY:Zackary DUNIVIN (University of Stuttgart) - "What is the role of social identity in LLMs?"
DESCRIPTION:Sociology Seminar \nTime: 10:00 am – 11:00 am\nDate: 25th of March\nRoom : 3049 \n  \nZackary DUNIVIN (University of Stuttgart) – “What is the role of social identity in LLMs?” \nAbstract :  \nFrontier language models don’t simply generate text. LLMs perform sociocultural sensemaking\, drawing on shared cultural patterns (e.g.\, roles\, norms\, social identity) to interpret situations and determine appropriate responses. This poses a governance problem. Identity can be necessary for ethical and competent judgment in context\, yet the dominant paradigm for regulating identity\, bias mitigation\, treats it mainly as a contaminant to suppress and evaluates systems largely via end outputs. That leaves underspecified how models should decide when identity is relevant\, how to reason under uncertainty and unequal power\, and how to justify and revise their stance. I call this problem bias negotiation: the procedural regulation of identity-conditioned judgments of relevance\, inference\, and justification. Bias negotiation matters for justice because a positive role for sociocultural reasoning is required to recognize and potentially remediate structural inequities. But it is equally implicated in core model functionality as sociocultural competence is critical in systems that operate across heterogeneous institutions and cultural contexts. Drawing on structured dialogues with multiple deployed chatbots\, I identify recurring negotiation repertoires and failure modes\, and conclude with specifications that support implementation and procedural evaluation design. \n  \nOrganizers: \nEtienne OLLION (Pôle sociologie CREST)\n \nPaola TUBARO (Pôle sociologie CREST) \nNicolas JULIA (Pôle sociologie CREST) \nPatrick PRÄG (Pôle sociologie CREST) \nSponsors:\nCREST \n
URL:https://crest.science/event/https-scholar-google-com-citationsuserqzadttcaaaajhlen/
CATEGORIES:Seminars,Sociology
ATTACH;FMTTYPE=:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20260325T121500
DTEND;TZID=Europe/Helsinki:20260325T133000
DTSTAMP:20260709T203938
CREATED:20250807T091358Z
LAST-MODIFIED:20260320T130748Z
UID:18284-1774440900-1774445400@crest.science
SUMMARY:Sophie BADE (Royal Holloway) - "Open-ended Matching with and without Markets"
DESCRIPTION:Séminaire Microéconomie : Tous les mercredis\nHeure : 12h15 – 13h30\nDate : 25/03/2026\nSalle : 3001 \nSophie BADE (Royal Holloway) – “Open-ended Matching with and without Markets” \nCV : Suppose agents are to be matched to objects and arrive over time without a definite terminal date. Although the set of core matchings can then be empty\, a transfinite version of the top trading algorithm shows that Pareto-optimal weak-core matchings always exist. Optimal matchings face a difficulty however: some of the agents linked by chains of trades may have lifespans that do not overlap\, thus obstructing their trades. To address this problem\, we let matchings be implemented via competitive markets. Competitive equilibria always exist and any matching in the core can be completely implemented. Moreover full core equivalence\, where allocations are in the core if and only if they can be completely implemented\, holds for a dense set of models. The extended algorithm also yields a strategyproof mechanism\, comparably to the finite model. \nOrganisateurs :  \nJulien COMBE (Pôle d’Economie du CREST)\n​​​​​​​​​​​​Yves Le YAOUANQ (Pôle d’Economie du CREST) \nCommanditaires :\nCREST \n
URL:https://crest.science/event/tommy-anderson-lund-u-tba/
CATEGORIES:Microeconomics,Seminars
ATTACH;FMTTYPE=:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20260325T140000
DTEND;TZID=Europe/Helsinki:20260325T150000
DTSTAMP:20260709T203938
CREATED:20260310T091602Z
LAST-MODIFIED:20260310T091743Z
UID:18856-1774447200-1774450800@crest.science
SUMMARY:Marshall TAYLOR (New Mexico State University) - "Meaningful Geometries: Towards a More Reliable Analysis of Culture using Word Embeddings"
DESCRIPTION:Sociology Seminar \nTime: 2:00 pm – 3:00 pm\nDate: 25th of March\nRoom : 3049 \n  \nMarshall TAYLOR “Meaningful Geometries: Towards a More Reliable Analysis of Culture using Word Embeddings” \nAbstract :  \nTo what extent do text corpora reflect gender or social class biases? In computational social science\, answering these types of questions often involves analyzing semantic dimensions in word embedding spaces. While powerful\, this approach relies on “anchor” words that lack standardized reliability assessments. I introduce a new\, structure-agnostic metric to fill this gap. I validate the metric at the word-level using expert- and crowd-sourced dictionaries and at the document-level using expert-annotated social media posts. I also provide simulation-based heuristic baselines to facilitate effect size interpretation and null hypothesis testing. \n  \nEtienne OLLION (Pôle sociologie CREST)\n \nPaola TUBARO (Pôle sociologie CREST) \nNicolas JULIA (Pôle sociologie CREST) \nPatrick PRÄG (Pôle sociologie CREST) \nSponsors:\nCREST \n
URL:https://crest.science/event/https-sociology-nmsu-edu-faculty-faculty-pages-drmarshalltaylor-html/
CATEGORIES:Seminars,Sociology
ATTACH;FMTTYPE=:
END:VEVENT
END:VCALENDAR