Cambridge ML Systems Seminar Series


2024-07-16 15:00: Navigating Privacy Risks in Language Models (Peter Kairouz -- Google)

2024-09-16 13:30: Energy Efficient Edge AI using Ensembles and Hyperdimensional Computing (Flavio Ponzina, UCSD)

2024-11-14 14:00: Physics in (Federated) Deep Neural Networks and Beyond: A Parametric Perspective (Zexi Li-Zhejiang University)

2024-11-26 14:00: Interpreting and Controlling Intermediate Representations in Large Language Models (Nicola Cancedda -Meta's Fundamental AI Research (FAIR) team. )

2024-11-29 11:00: Balancing Quality and Efficiency in Future AI Systems (Shiqiang Wang - IBM)

2025-01-28 15:00: Training LLMs Anywhere: Enabling Large-Scale Decentralized Learning on Your Mobiles Devices (Dongqi Cai - Beijing University of Posts and Telecommunications)

2025-02-25 16:05: Towards the Responsible Development of Foundation Model and Generative AI (Dr. Lingjuan Lyu )

2025-04-08 14:00: Learning Rate Schedules, Scaling Laws, and Techniques for Pretraining LLMs (Alex Hägele, EPFL)

2025-05-02 14:00:  Frontiers in Embodied AI for Autonomous Driving (Jamie Shotton, Wayve)

2025-05-20 15:30: Building Organizational Resilience with AI Malfunction Drills (Rosco Hunter, University of Warwick )

2025-06-18 15:00: Drawing Connections in Decentralized Deep Learning (Max Ryabinin)

2025-08-04 11:00: Learning Under Constraints: From Federated Collaboration to Black-Box LLMs (Salma Kharrat, Kaust)

2025-10-20 11:00: Federated Learning at H.IAAC: On-going Research and Opportunities (Allan M. de Souza & Luiz Bittencourt, Universidade Estadual de Campinas (UNICAMP), Brazil)

2025-10-22 11:00: Client Clustering for Federated Learning in Data Heterogeneous Scenarios (Gabriel Ukstin Talasso, Campinas (Unicamp), Brazil)

2025-11-07 11:00: DISCO: Dynamical Integration Systems for Convergence Optimisation in Distributed Low-Communication Training (Finn Anderson )

2026-02-16 14:30: Bringing distributed training natively to Transformers library (Ferdinand Mom - Hugging Face)

2026-03-09 13:00: Learning to Discover Learning Algorithms (Alex Goldie - University of Oxford )