Generative AI in Data Science and Analysis

Dr. Thanachart Numnonda
Dr. Thanachart Numnonda
Profile : https://shorturl.at/lWuXj
Abstract

Generative AI in data science and analysis, focusing on tools like ChatGPT, Gemini, and Claude.ai. It aims to equip participants with the skills to employ AI technologies in extracting, interpreting, and analyzing data without the need for traditional programming. Attendees will explore the functionalities of ChatGPT across different versions, gaining insights into how these tools can be used for deep data analysis and decision-making processes in various data science contexts.

The workshop will delve into practical aspects of data science, including data engineering, visualization, predictive analytics, and natural language processing, all through the lens of Generative AI. Participants will engage in hands-on lab sessions, applying these AI tools to real-world data science problems, and learning to conduct end-to-end data projects. This program is designed for individuals looking to integrate advanced AI capabilities into their data science toolkit, enabling them to efficiently handle complex data analysis and predictive modeling tasks with minimal reliance on coding or external IT resources.

Biography

Dr. Thanachart Numnonda is an executive director of IMC Institute. He has a diverse and extensive professional background. Currently, he holds several key positions, including being an independent director and chairman of risk committees in various companies like Thanachart Capital Limited, SiamEast Solutions public company Limited, VinTcom Technology Public Company Limited, and Humanica public company Limited. He is also involved in academia, serving as a council member and chairman of risk committees in several public universities.

Quantum Intelligence : An Introduction to Quantum Computing and Quantum Machine Learning

Dr. Thanachart Numnonda
Prof. Dr. Stephen John Turner
Profile : https://vistec.ac.th/stephen-john-turner
Abstract

Quantum computing is the study of information processing based on the quantum properties of matter. By combining the rich representational power of quantum states with the possibility of exponential parallelism, quantum computing has the potential to revolutionize many aspects of science, technology and industry. While quantum computing is still at an early stage of development, we have recently seen rapid advances in quantum technology, and quantum computers with hundreds of qubits are now readily available via cloud services. In this current “utility” era, quantum computers are demonstrating their potential to provide an advantage over classical computers in certain application areas.

The first part of this tutorial will explain the basic concepts of quantum systems and their properties, including superposition, entanglement and interference, and how the inherent parallelism of quantum computers may be exploited. Examples of quantum algorithms and applications will be given, including quantum networks and security. The second part of the tutorial will describe approaches to quantum machine learning (QML). Some important QML algorithms will be described, including quantum neural networks, with examples taken from real-world applications. Finally, the tutorial will explore the exciting possibilities offered by the convergence of quantum computing and artificial intelligence to form quantum intelligence.

Biography

Stephen John Turner is Professorial Fellow in Quantum Computing at Vidyasirimedhi Institute of Science and Technology (Thailand). From 2008-2015, he was full Professor of Computer Science at Nanyang Technological University (Singapore), having joined the University as an Associate Professor in 2000. During his time there, he was Director of the Parallel and Distributed Computing Centre and subsequently Head of the Networks and Distributed Systems Division in the School of Computer Engineering.

He received his MA in Mathematics and Computer Science from Cambridge University (UK) and his MSc and PhD in Computer Science from Manchester University (UK). He is a Chartered IT Professional and Chartered Engineer (UK). His main research interests are: Quantum Computing, Simulation and Optimization, Complex Systems, Internet of Things, and Edge and Cloud Computing. He has published extensively and has received a number of best paper awards, particularly for his work in Parallel and Distributed Simulation.

An introduction to training and optimizing Large Language Models

Dr. Thanachart Numnonda
Asst. Prof. Dr. Jan N. van Rijn
Profile : https://www.universiteitleiden.nl
Abstract

Transformer-based language models have achieved milestones in natural language processing, but they come with challenges, mainly due to their computational footprint. While large language models are readily available for use, it remains important to do academic research towards these for the following reasons:

(i) the training procedure and datasets of these models are typically not disclosed, and they are essentially operating as black-boxes maintained by companies.

(ii) the size of these models gives them a high computational footprint, even when deployed (inference-stage). Therefore, developing compute-efficient models that can be deployed with the computational resources available at the disposal of small- and medium-size enterprises is crucial.

(iii) models such as ChatGPT interact with society in ways that we could not imagine several years before, but many open questions remain, related to trustworthiness, privacy, security, and efficiency, especially in lower-resource contexts. It is therefore important that the research community addresses these topics and develops open-source models for future applications.

In this tutorial, we will present our research regarding how to train and optimise large language models. It will cover both the pre-training as well as the finetuning stage.

Biography

Jan N. van Rijn holds a tenured position as assistant professor at Leiden University, where he works in the computer science department (LIACS) and Automated Design of Algorithms cluster (ADA). His research interests include trustworthy artificial intelligence, automated machine learning (AutoML) and metalearning. He obtained his PhD in Computer Science in 2016 at Leiden Institute of Advanced Computer Science (LIACS), Leiden University (the Netherlands). During his PhD, he developed OpenML.org, an open science platform for machine learning, enabling sharing of machine learning results. He made several funded research visits to the University of Waikato (New Zealand) and the University of Porto (Portugal). After obtaining his PhD, he worked as a postdoctoral researcher in the Machine Learning lab at the University of Freiburg (Germany), headed by Prof. Dr. Frank Hutter, after which he moved to work as a postdoctoral researcher at Columbia University in the City of New York (USA). In 2023, he visited the College of Computing, Khon Kaen University, Thailand. His research aim is to democratize access to machine learning and artificial intelligence across societal institutions, by developing knowledge and tools that support domain experts, and make AI-experts more aware of safety risks. He is one of the authors of the book ‘Metalearning: Applications to Automated Machine Learning and Data Mining’ (freely accessible, published by Springer).