Mapping Hong Kong’s Financial Ecosystem: With Networks and LLMs

We present the first study of the Public Register of Licensed Persons and Registered Institutions maintained by the Hong Kong Securities and Futures Commission (SFC) through the lens of complex network analysis. This dataset, spanning 21 years with daily granularity, provides a unique view of the evolving social network between licensed professionals and their affiliated firms in Hong Kong’s financial sector. Leveraging large language models, we classify firms (e.g., asset managers, banks) and infer the likely nationality and gender of employees based on their names. This application enhances the dataset by adding rich demographic and organizational context, enabling more precise network analysis. Our preliminary findings reveal key structural features, offering new insights into the dynamics of Hong Kong’s financial landscape. We release the structured dataset to enable further research, establishing a foundation for future studies that may inform recruitment strategies, policy-making, and risk management in the financial industry.

Speaker:

Abdulla Alketbi - Research - Specialist

Unveiling the Black Box: A Guide to Explainable and Interpretable ML

As machine learning models become more integral to decision-making processes, understanding their inner workings is crucial. This presentation delves into the concepts of explainable and interpretable ML, providing an overview of how complex “black box” models can be made more transparent. Drawing from model-agnostic techniques we explore approaches to ensure that ML systems are not only powerful but also understandable. We’ll highlight methods like Partial Dependence Plots, Marginal Plots, ALE and SHAP values, all aimed at balancing model performance with clarity. The goal is to equip participants with practical tools and insights for applying interpretable machine learning in real-world contexts without sacrificing predictive accuracy. Based on Christoph Molnar’s comprehensive work on interpretable machine learning, this session provides both a theoretical foundation and practical examples for enhancing ML transparency.

Speaker:

Kristof Juhasz - Quantitative - Researcher

TelecomGPT: Revolutionizing Telecommunications with Large Language Models

TelecomGPT has recently made the headlines (see UAE's TelecomGPT: The AI Breakthrough Set to Transform Telecoms) as the first general-purpose Large Language Models (LLMs) for specialized applications within the telecommunications domain. The work was jointly performed by Khalifa University's (KU) 6G Center and Abu Dhabi's Technology Innovation Institute (TII). Leveraging customized pre-training, instruction tuning, and alignment techniques, TelecomGPT addresses unique challenges in telecom, including domain-specific language understanding, code generation, and real-time network optimization. In this talk, we will delve into the framework's architecture, key benchmarks, and its potential to enhance network management, reduce operational costs, and drive new AI-driven innovations in telecom. The discussion will highlight TelecomGPT's role in transforming the telecommunications landscape, offering unprecedented opportunities for industry advancement and efficiency.

Speaker:

Prof. Merouane Debbah - Professor - Khalifa University