In person

Edge AI in action: Accelerating development across the Global South

  • Date
    20 February 2026
    Timeframe
    13:30 - 14:25 IST
    Duration
    55 minutes

      As AI evolves, the shift from centralized cloud computing to edge AI is transforming how data is processed and acted upon. By moving intelligence closer to data sources such as sensors, mobile devices, and IoT systems, edge AI reduces latency, bandwidth usage, and energy consumption while improving real-time decision-making, privacy, and security. Real-world applications will be highlighted along with emerging lightweight and generative AI models designed for resource-constrained edge environments.

      Agenda

      As AI moves from the cloud to the edge, it starts making decisions closer to where sensing, communication, and action actually happen. This shift changes how intelligent systems behave—especially when latency, uncertainty, and limited resources come into play. In this talk, Prof. Lall will look at Edge AI not just as a deployment choice, but as a change in how large systems are designed and controlled. He will discuss why coordination, robustness, and predictable behavior become critical when intelligence is spread across many interacting components, and why system-level thinking is essential for Edge AI to work reliably at scale, particularly in complex settings like India.

      Federated Learning is emerging as a key enabler of Edge Artificial Intelligence in 6G networks, ensuring distributed and privacy-preserving model training across devices, base stations, and edge servers without the need to share raw data with a central server. By supporting hierarchical and communication-efficient aggregation, FL helps realize AI-native wireless control for applications such as adaptive radio resource management, intelligent network slicing, mobility prediction, and collaborative security. This talk will discuss the architecture of federated edge learning in 6G and highlight practical use cases demonstrating how distributed intelligence can enable scalable, low-latency, and privacy-aware next-generation networks.

      Are you sure you want to remove this speaker?