Overview
The NeurIPS 2019 workshop titled “Bridging Game Theory and Deep Learning” aims to explore the intersection of these two fields. With the rise of generative modeling and adversarial learning, there is a growing interest in differentiable two-player games, particularly in the context of generative adversarial networks (GANs). This workshop seeks to address the unique challenges posed by these games, which differ from standard minimization tasks typically encountered in machine learning (ML). The workshop serves as a platform for researchers to discuss theoretical and practical aspects of smooth games and their applications in ML.
Background & Relevance
Game theory and deep learning are increasingly intertwined, especially as ML practitioners seek to apply optimization techniques to game-theoretic problems. The previous workshop held at NeurIPS 2018 focused on theoretical aspects and dynamics of games in ML, highlighting fundamental topics and challenges. This year’s workshop aims to build on that foundation, inviting contributions that explore open questions and innovative applications at the intersection of these domains. The relevance of this workshop is underscored by the growing need for effective algorithms and models that can handle complex interactions in learning contexts.
Key Details
- Event: NeurIPS 2019 Workshop on Bridging Game Theory and Deep Learning
- Dates: December 13-14, 2019
- Location: Vancouver
- Submission Deadline: September 16, 2019 (11:59 PM AoE)
- Acceptance Notification: September 30, 2019
- Camera Ready Submission: November 15, 2019 (11:59 PM AoE)
- Website: Smooth Games Workshop
Eligibility & Participation
This workshop targets both theoretical and applied researchers from the fields of ML, mathematical programming, and game theory. Participants are encouraged to submit contributions that address specific questions related to the integration of learning and game theory, as well as practical implications and applications of these concepts.
Submission or Application Guidelines
Contributors should submit an anonymous extended abstract of 2-4 pages (excluding references) formatted according to the modified NeurIPS style. Submissions must be made via CMT. The workshop emphasizes topic relevance as a primary criterion for acceptance.
Additional Context / Real-World Relevance
The integration of game theory and deep learning is crucial for advancing our understanding of interactions in multi-agent systems. As AI systems become more complex, the ability to model and analyze these interactions will be essential for developing robust algorithms and applications. This workshop provides a timely opportunity for researchers to share insights and collaborate on addressing the challenges at this intersection.
Conclusion
Researchers and practitioners in the fields of game theory and deep learning are encouraged to participate in this workshop. By contributing your work, you can help shape the future of research at this exciting intersection. For more information, visit the workshop website and prepare your submissions ahead of the deadline.
Category: CFP & Deadlines
Tags: game-theory, deep-learning, neurips, machine-learning, optimization, generative-adversarial-networks, smooth-games, mathematical-programming
