Location: Summit Convention Center 900 Pine St, Seattle, WA 98101
Main room: Room 420-421 Breakout rooms: Room 422-425
DM Workshop Co-Chairs
Shixiang (Woody) Zhu, Carnegie Mellon University
Andi Wang, University of Wisconsin-Madison
Hadis Anahideh, University of Illinois Chicago
Program Schedule
DMDA Workshop Schedule.pdfKeynote Speakers
H. Milton Stewart School of Industrial and Systems Engineering
Georgia Institute of Technology
Generative models for high-dimensional statistical inference
Yao Xie, Georgia Institute of Technology
Generative AI (GenAI), with its ability to synthesize and generate data, offers great potential for addressing high-dimensional data challenges. While generative models have advanced considerably, their full impact on high-dimensional statistical inference remains underexplored. We view generative models as neural network-based representations of continuous probability density functions learned from data and propose a framework that adapts the flow-based generative models for statistical inference and decision-making under uncertainty. Our approach moves beyond merely generating data resembling observations, emphasizing instead the mapping of a reference measure to a target distribution—whether defined by samples, an optimization solution, or a parametric form up to an unknown constant. We demonstrate the utility of this framework for density estimation, distributionally robust optimization (DRO), posterior sampling, and importance sampling.
Learning and discrete optimization with irreversible decisions
Alexandre Jacquillat, Massachusetts Institute of Technology
This talk presents several modeling and algorithmic approaches combining machine learning and mixed-integer optimization within multi-stage decision-making under uncertainty. The first part focuses on learning-then-optimize algorithms to support resource allocation. Examples are drawn from transportation (learning passenger itineraries to optimize network-wide air traffic management) and healthcare (learning the dynamics of a pandemic to optimize vaccine allocations). The second part focuses on online optimization, by proposing a single-sample online approximation (SSOA) approach that approximates unknown parameters with a single realization of uncertainty and re-optimizes decisions dynamically over a rolling horizon. This algorithm is applied to cloud supply chain operations, leading to improved power utilization, significant cost savings, and concomitant reductions in greenhouse gas emissions. The third part combines learning and optimization in online covering: at each period, a decision-maker selects facilities, receives information on the success of each one, and updates a machine learning model to guide future decisions in order to achieve a coverage target. We find that the regret grows sub-linearly in an asymptotic regime characterized by a large coverage target, and that the regret rate converges exponentially fast as the planning horizon grows larger. These findings underscore the benefits of limited online learning and optimization, in that even a few rounds can provide significant benefits as compared to a no-learning baseline.
Sponsored by
Platinum Level
Gold Level
Silver Level
Call for Papers
The 19th INFORMS Workshop on Data Mining and Decision Analytics
October 19, 2024, Seattle, WA
Organized by INFORMS Data Mining (DM) Society
The Data Mining Society of INFORMS is organizing the 19th INFORMS Workshop on Data Mining and Decision Analytics on Oct 19, 2024, the day before the 2024 INFORMS Annual Meeting. You are cordially invited to attend the workshop and present your recent research on data mining, decision analytics, and artificial intelligence. To participate in the Workshop, a full paper must be submitted by the deadline of August 8, 2024, for consideration. All accepted papers are automatically considered for the DMDA Workshop best paper competition, either in the theoretical track or in the applied research tracks, according to the presenters’ specifications. The winner and runner-ups will be announced at the Workshop and recommended for fast-track submission to the INFORMS Journal on Data Science (IJDS).
Topics of interest include, but are not limited to:
Applied Data Analytics
Optimization and Decision-Making
Simulation Network, Graph, and Web Analysis
Bayesian Data Analytics
Longitudinal Data Analysis
Causal Inference
Anomaly Detection
Reliability & Maintenance
Ethics, Privacy, Security, and Fairness
Text Mining & Natural Language Processing
Generative AI & Large Language Models
Timeline
● June 1: Paper submission begin
● August 8: Paper submission close
● September 1: Final review decision
● September 22: Workshop on Data Mining and Decision Analytics registration deadline
We apologize for the delay, we will finalize everything asap.
Papers submission guideline
● Maximum of 10 pages (including abstract, tables, figures, and references).
● Single-spacing and 11-point font with one-inch margins on four sides.
● Papers must be submitted via this link. Late submission will not be considered for further review.
● Copyright: The DM workshop will not retain the copyrights on the papers; so, the authors are free to submit their papers to other outlets.
DM Workshop Co-Chairs
● Shixiang (Woody) Zhu (shixianz@andrew.cmu.edu), Carnegie Mellon University
● Andi Wang (andi_wang@yahoo.com), University of Wisconsin-Madison
● Hadis Anahideh (hadis@uic.edu), University of Illinois Chicago
Registration
Onsite only
$75 for students and retirees
$150 for professionals
$150 for members
Previous Workshops
18th INFORMS Workshop on Data Mining and Decision Analytics
17th INFORMS Workshop on Data Mining and Decision Analytics
16th Hybrid INFORMS Workshop on Data Mining and Decision Analytics
15th Virtual INFORMS Workshop on Data Mining and Decision Analytics
Please contact us at dmda2024@gmail.com if you have any questions.
© 2024 INFORMS Workshop on Data Mining and Data Analytics. All rights reserved.
