Site menu:
OPT2025
We welcome you to participate in the 17th International OPT Workshop on Optimization for Machine Learning, to be held as a part of the NeurIPS 2025 conference. This year we particularly encourage (but not limit) submissions with a focus on "Statistics Meets Optimization".
We are looking forward to an exciting OPT!
Call for Participation
Important Dates
TBDInvited Talks
TBDOverview
Optimization lies at the heart of many machine learning algorithms and enjoys great interest in our community. Indeed, this intimate relation of optimization with ML is the key motivation for the OPT series of workshops. We aim to foster discussion, discovery, and dissemination of state-of-the-art research in optimization relevant to ML.
We invite participation in the 17th International (in-person) Workshop on "Optimization for Machine Learning", to be held as a part of the NeurIPS 2025 conference. We invite high quality submissions for presentation as spotlights or poster presentations during the workshop. We are especially interested in participants who can contribute theory / algorithms, applications, or implementations with a machine learning focus and encourage work-in-progress and state-of-art ideas.
All accepted contributions will be listed on the workshop webpage and are expected to be presented as a poster during the workshop. A few submissions will in addition be selected for contributed talks or for short spotlight presentations.
We particularly encourage submissions in the area of “Statistics Meets Optimization.” Since its inception, stochastic optimization has been grounded in statistical principles, and today, many challenges in machine learning—such as understanding generalization in overparameterized models, characterizing the training dynamics of large models, and developing algorithms that adapt to data structure—are tightly linked to statistical concepts. OPT 2025 aims to bridge statistical and optimization perspectives in addressing these challenges, especially in the context of large-scale models and modern ML practice.
The main topics are, including, but not limited to:
- Adaptive Stochastic Methods
- Algorithms and techniques (higher-order methods, algorithms for nonsmooth problems, optimization with sparsity constraints, online optimization, streaming algorithms)
- Approaches to Adversarial Machine Learning
- Average-case Analysis of Optimization Algorithms
- Combinatorial optimization for machine learning
- Deep learning optimization
- Federated learning
- Games; min/max theory
- Nonconvex Optimization
- Optimization software (integration with existing DL software, hardware accelerators and systems)
- Parallel and Distributed Optimization for large-scale learning
- Privacy and Optimization
- Scaling laws
- The Interface of Generalization and Optimization
Submission Instructions:
TBDLooking forward to another great OPT workshop!
The Organizing Committee:
- Misha Belkin (University of California San Diego)
- Cristóbal Guzmán (chair) (Pontificia Universidad Católica de Chile)
- Frederik Kunstner (INRIA)
- Zak Mhammedi (MIT)
- Courtney Paquette (McGill University)
- Sebastian Stich (CISPA Helmholtz Center for Information Security)