BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//DDSA - ECPv6.16.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:DDSA
X-ORIGINAL-URL:https://ddsa.dk
X-WR-CALDESC:Events for DDSA
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Europe/Copenhagen
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20230326T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20231029T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20240331T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20241027T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20250330T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20251026T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240506T163000
DTEND;TZID=Europe/Copenhagen:20240506T173000
DTSTAMP:20260531T033146
CREATED:20240209T142824Z
LAST-MODIFIED:20240209T142824Z
UID:10000362-1715013000-1715016600@ddsa.dk
SUMMARY:Online learning and decision-making for renewables participating in electricity markets
DESCRIPTION:This is a talk given by Prof Pierre Pinson as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/\, branding the role of Operations Research in Artificial Intelligence with the support of EURO\, https://www.euro-online.org. \nAbstract: \nThere is extensive literature on the analytics involved in the participation of renewable energy producers in electricity markets\, covering both forecasting and decision-making. In their simplest form\, participation strategies are to be seen as newsvendor problems (taking a decision-making perspective)\, or quantile regression problems (if taking a forecasting perspective instead). We will therefore explore recent advances at the interface between learning\, forecasting and stochastic optimisation of relevance to renewable energy producers participating in electricity markets. This will cover online learning and decision-making\, as well as distributionally robust optimisation. \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/c4046bb5460245e48247284d998a25c2 \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/online-learning-and-decision-making-for-renewables-participating-in-electricity-markets/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240422T163000
DTEND;TZID=Europe/Copenhagen:20240422T173000
DTSTAMP:20260531T033146
CREATED:20240209T113011Z
LAST-MODIFIED:20240209T113011Z
UID:10000361-1713803400-1713807000@ddsa.dk
SUMMARY:Integrating Stochastic Optimization and Machine Learning via Residuals
DESCRIPTION:This is a talk given by Prof Güzin Bayraksan as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/\, branding the role of Operations Research in Artificial Intelligence with the support of EURO\, https://www.euro-online.org. \nAbstract: \nWe consider data-driven approaches that integrate a machine learning prediction model within stochastic optimization\, given joint observations of uncertain parameters and covariates. Given a new covariate observation\, the goal is to choose a decision that minimizes the expected cost conditioned on this observation. We first present a Sample Average Approximation (SAA) approach for approximating this problem that incorporates residuals from the learning step. Then\, in the limited-data regime\, we consider Distributionally Robust Optimization (DRO) variants of these models. Our framework is flexible in the sense that it can accommodate a variety of learning setups and DRO ambiguity sets. We investigate the asymptotic and finite sample properties of SAA and the DRO variants obtained using Wasserstein\, sample robust optimization\, and phi-divergence-based ambiguity sets. We discuss extensions to decision-dependent settings and present applications using real-world data. \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/1b732023cea64e6da05b1e0d4d587578 \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/integrating-stochastic-optimization-and-machine-learning-via-residuals/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240408T163000
DTEND;TZID=Europe/Copenhagen:20240408T173000
DTSTAMP:20260531T033146
CREATED:20240209T142945Z
LAST-MODIFIED:20240209T142945Z
UID:10000364-1712593800-1712597400@ddsa.dk
SUMMARY:YOUNG Machine Learning NeEDS Mathematical Optimization Seminar
DESCRIPTION:This seminar consists of three talks\, as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/ \nSpeaker: Cheng Peng\nTitle: Factor Model of Mixtures \nSpeaker: Haofeng Zhang\nTitle: Estimate-Then-Optimize versus Integrated-Estimation-Optimization versus Sample Average Approximation: A Stochastic Dominance Perspective \nSpeaker: Giulia Di Teodoro\nTitle: Unboxing Tree Ensembles for interpretability: a hierarchical visualization tool and a multivariate optimal re-built tree \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/3ef5ca638bf54c85aee046eb54559d2e \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/young-machine-learning-needs-mathematical-optimization-seminar-4/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240318T163000
DTEND;TZID=Europe/Copenhagen:20240318T173000
DTSTAMP:20260531T033146
CREATED:20240209T112857Z
LAST-MODIFIED:20240209T112857Z
UID:10000360-1710779400-1710783000@ddsa.dk
SUMMARY:FunSearch: Discovering new mathematics and algorithms using Large Language Models
DESCRIPTION:This is a talk given by Dr Bernardino Romera Paredes as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/\, branding the role of Operations Research in Artificial Intelligence with the support of EURO\, https://www.euro-online.org. \nAbstract: \nIn this talk I will present FunSearch\, a method to search for new solutions in mathematics and computer science. FunSearch works by pairing a pre-trained LLM\, whose goal is to provide creative solutions in the form of computer code\, with an automated “evaluator”\, which guards against hallucinations and incorrect ideas. By iterating back-and-forth between these two components\, initial solutions “evolve” into new knowledge. I will present the application of FunSearch to a central problem in extremal combinatorics — the cap set problem — where we discover new constructions of large cap sets going beyond the best known ones\, both in finite dimensional and asymptotic cases. This represents the first discoveries made for established open problems using LLMs. Then\, I will present the application of FunSearch to an algorithmic problem\, online bin packing\, which showcases the generality of the method. In this use case\, FunSearch finds new heuristics that improve upon widely used baselines. I will conclude the talk by discussing the implications of searching in the space of code. \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/5cd1f5545f6642939131e5035de434d3 \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/funsearch-discovering-new-mathematics-and-algorithms-using-large-language-models/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240311T163000
DTEND;TZID=Europe/Copenhagen:20240311T173000
DTSTAMP:20260531T033146
CREATED:20240209T112651Z
LAST-MODIFIED:20240226T080447Z
UID:10000359-1710174600-1710178200@ddsa.dk
SUMMARY:Role of Analytics Professionals in Building a Data Driven Culture
DESCRIPTION:This is a talk given by Dr Pooja Dewan as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/\, branding the role of Operations Research in Artificial Intelligence with the support of EURO\, https://www.euro-online.org. \nAbstract: \nThis talk is going to focus on the value of Data Ecosystems\, Starting with the business outcome\, Composition of teams\, and evolving skills needed to take Algorithms in R&D to Commercial Reality and building a Data Driven Culture and creating impact. \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/8d60cb95de7e431e83be6381f82e4e23 \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/role-of-analytics-professionals-in-building-a-data-driven-culture/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240304T163000
DTEND;TZID=Europe/Copenhagen:20240304T173000
DTSTAMP:20260531T033146
CREATED:20240209T142909Z
LAST-MODIFIED:20240209T142909Z
UID:10000363-1709569800-1709573400@ddsa.dk
SUMMARY:YOUNG Machine Learning NeEDS Mathematical Optimization Seminar
DESCRIPTION:This seminar consists of three talks\, as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/ \nSpeaker: Nathan Justin\nTitle: Learning Optimal Classification Trees Robust to Distribution Shifts \nSpeaker: José Ángel Martín Baos\nTitle: Can machine learning methods effectively model travel mode choice? Beyond predictive performance \nSpeaker: Ioana Molan\nTitle: Learning the Follower’s Objective Function in Sequential Bilevel Games \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/f36b823fbfc74849848d66808d8db459 \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/young-machine-learning-needs-mathematical-optimization-seminar-3/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240226T163000
DTEND;TZID=Europe/Copenhagen:20240226T173000
DTSTAMP:20260531T033146
CREATED:20240209T112600Z
LAST-MODIFIED:20240209T112600Z
UID:10000358-1708965000-1708968600@ddsa.dk
SUMMARY:Learning and Optimization: Separate or Integrate?
DESCRIPTION:This is a talk given by Prof Nathan Kallus as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/\, branding the role of Operations Research in Artificial Intelligence with the support of EURO\, https://www.euro-online.org. \nAbstract: \nPredictive side information (aka context) makes optimization under uncertainty less uncertain\, but leveraging it requires we learn a potentially complex predictive relationship. We might use off-the-shelf ML methods for this\, but their training process ignores the downstream optimization task where the model would be plugged in. Instead\, we can train the model in an end-to-end fashion to optimize the downstream costs of the decisions it would induce. In this talk I will tackle the question\, which is better? That is\, should we separate or integrate the learning and optimization tasks? I show that in linear problems\, where we only care to learn the mean-prediction (aka regression) function\, the naive separated approach surprisingly has regret that vanishes orders faster than end-to-end methods — a consequence of real problems not having arbitrarily bad near-dual-degeneracy. However\, for general (nonlinear) contextual optimization\, which involves the hard-to-learn conditional-probability-prediction function\, it may be better to bypass learning this complex object by integrating the tasks and directly learning a decision policy. I show how to do this tractably near-optimally for tree-based policies and build scalable decision forests using approximate splitting criteria based on second-order optimization perturbation analysis. \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/5f6b2827c847432ab35034deb752df08 \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/learning-and-optimization-separate-or-integrate/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240219T163000
DTEND;TZID=Europe/Copenhagen:20240219T173000
DTSTAMP:20260531T033146
CREATED:20240209T112521Z
LAST-MODIFIED:20240209T112521Z
UID:10000357-1708360200-1708363800@ddsa.dk
SUMMARY:Risk Quadrangle and Applications: Support Vector Regression (SVR)
DESCRIPTION:This is a talk given by Prof Stan Uryasev as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/\, branding the role of Operations Research in Artificial Intelligence with the support of EURO\, https://www.euro-online.org. \nAbstract: \nThe Support Vector Regression (SVR) is investigated in the Risk Quadrangle framework. ε-SVR and ν-SVR\, correspond to the minimization of equivalent error measures (Vapnik error and CVaR norm) with a regularization. These two error measures\, define corresponding Risk Quadrangles. We show that SVR is an asymptotically unbiased estimator of the average of two symmetric conditional quantiles. Equivalence of ε-SVR and ν-SVR follows from the quadrangle framework. SVR is formulated as a deviation minimization with a regularization penalty by applying Error Shaping Decomposition of Regression. Moreover\, f-Divergence Quadrangle implies that SVR is a robust variant of the mean-absolute regression with regularization. Finally\, the dual formulation of SVR is derived. \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/a4b0f61d3ca245c28f1a4b183a48b800 \nOrganizers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/risk-quadrangle-and-applications-support-vector-regression-svr/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/Copenhagen:20240212T163000
DTEND;TZID=Europe/Copenhagen:20240212T173000
DTSTAMP:20260531T033146
CREATED:20240209T112422Z
LAST-MODIFIED:20240209T112422Z
UID:10000356-1707755400-1707759000@ddsa.dk
SUMMARY:Decentralized Bilevel Optimization
DESCRIPTION:This is a talk given by Prof Shiqian Ma as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\, https://congreso.us.es/mlneedsmo/\, branding the role of Operations Research in Artificial Intelligence with the support of EURO\, https://www.euro-online.org. \nAbstract: \nBilevel optimization has received tremendous attention recently due to its great success in machine learning applications such as meta learning\, reinforcement learning\, and hyperparameter optimization. In this talk\, we discuss recent advances on bilevel optimization algorithms in decentralized networks. We will address the issues of how to estimate the hypergradient in decentralized networks\, how to design single-loop algorithms\, and how to improve the per-iteration complexity using moving average techniques\, etc. \nMore information \nLink to the talk: https://eu.bbcollab.com/guest/16aa78a6789c46c39744a51293653b13 \nOrganisers \nThe organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization\nEmilio Carrizosa\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nNuria Gómez-Vargas\, IMUS-Instituto de Matemáticas de la Universidad de Sevilla\nThomas Halskov\, Copenhagen Business School\nDolores Romero Morales\, Copenhagen Business School
URL:https://ddsa.dk/event/decentralized-bilevel-optimization/
LOCATION:Online
CATEGORIES:Other Events
END:VEVENT
END:VCALENDAR