Upcoming Machine Learning and AI Seminars: April 2025 Edition

This guide aims to inform a global audience about these free virtual events, providing insights into each seminar's focus, speakers, and registration details.

April 2025 AI Seminars

1. April 1, 2025

   • Lie-Poisson Neural Networks (LPNets): Data-Based Computing of Hamiltonian Systems

     • Speaker: Vakhtang Poutkaradze from the University of Alberta

     • Organized by: University of Minnesota

     • Overview: This seminar delves into Lie-Poisson Neural Networks (LPNets), emphasizing their application in computing Hamiltonian systems through data-driven methods.

     • Registration: Interested participants can register via the University of Minnesota's seminar page.

   • Sample Complexity of Data-Driven Tuning of Model Hyperparameters in Neural Networks with Structured Parameter-Dependent Dual Function

     • Speaker: Anh Nguyen from Carnegie Mellon University

     • Organized by: Carnegie Mellon University

     • Overview: This talk focuses on the intricacies of tuning model hyperparameters in neural networks, particularly examining sample complexity and structured parameter-dependent dual functions.

     • Registration: Access the seminar through Carnegie Mellon University's provided Zoom link.

2. April 3, 2025

   • Optimizing Green Infrastructure for Climate-Resilient Cities

     • Speaker: Abdul Shaamala from Queensland University of Technology

     • Organized by: Research Institutes of Sweden (RISE)

     • Overview: This seminar explores strategies for enhancing urban green infrastructure to bolster climate resilience in metropolitan areas.

     • Registration: Participants can register through RISE's official website.

3. April 7, 2025

   • Harnessing Transport Theory for Post-Hoc Explanations of Machine Learning

     • Speaker: Lei You

     • Organized by: Imperial College London

     • Overview: The session discusses the application of transport theory in providing post-hoc explanations for machine learning models, aiming to enhance interpretability.

     • Registration: Sign up via Imperial College London's seminar registration page.

   • Deep Generative Models in Stochastic Optimization

     • Speaker: David Pisinger from the Technical University of Denmark

     • Organized by: Association of European Operational Research Societies

     • Overview: This talk examines the role of deep generative models in stochastic optimization, highlighting advancements and practical applications.

     • Registration: Join the mailing list to receive the seminar link.

4. April 8, 2025

   • AI for Science Series

     • Speaker: Oliver Dicks from the University of British Columbia

     • Organized by: École Polytechnique Fédérale de Lausanne (EPFL)

     • Overview: Part of EPFL's AI for Science series, this seminar showcases the integration of artificial intelligence in scientific research and discovery.

     • Registration: Access the Zoom link through EPFL's event page.

   • On Provable Posterior Sampling with Denoising Oracles

     • Speaker: Joan Bruna from NYU Courant Institute

     • Organized by: University of Minnesota

     • Overview: The seminar addresses posterior sampling techniques utilizing denoising oracles, focusing on theoretical guarantees and applications.

     • Registration: Register via the University of Minnesota's seminar portal.

   • Situating High School Data Science in the Lives of Students

     • Speakers: David Weintrop, Rotem Israel-Fishelson, and Peter F. Moon from the University of Maryland

     • Organized by: Raspberry Pi Foundation

     • Overview: This session explores the integration of data science education in high schools, emphasizing relevance to students' daily lives.

     • Registration: Sign up through the Raspberry Pi Foundation's event page.

5. April 10, 2025

   • Controlling Diffusion Models at Inference Time

     • Speaker: Kirill Neklyudov from MILA and Université de Montréal

     • Organized by: Chalmers AI4Science

     • Overview: The talk focuses on techniques for controlling diffusion models during inference, enhancing model performance and reliability.

     • Registration: Access the seminar via Chalmers AI4Science's Zoom link; use password: ai4science.

   • Challenges for AI in Multimodal STEM Assessments: A Human-AI Comparison

     • Speaker: Anna Sotnikova

     • Organized by: École Polytechnique Fédérale de Lausanne (EPFL)

     • Overview: This seminar examines the challenges faced by AI in multimodal STEM assessments, comparing human and AI performance.

     • Registration: Register through EPFL's event registration page.

6. April 15, 2025

   • Federated Reinforcement Learning: Statistical, Communication, and Computation Trade-offs

     • Speaker: Yuejie Chi from Carnegie Mellon University

     • Organized by: University of Minnesota

     • Overview: The session delves into federated reinforcement learning, discussing the balance between statistical efficiency, communication overhead, and computational complexity.

     • Registration: Register via the University of Minnesota's seminar page.