Seonglae ChoWhy are you applying for this specific programme, and how does it fit with your future plans?
My genuine interest and established goal in Artificial Intelligence(AI) firmly stem from my intellectual curiosity and extensive hands-on experience. I am driven to uncover how LLM (Large Language Model) functions, which led me to investigate Explainable AI, specifically Mechanistic Interpretability. This interest guided my research to explore interpretable decision-making within AI summarization which are accepted from NAACL 2024. Also, my three years of working experience in autonomous driving technology sector reinforced my belief in the necessity for a deep understanding to ensure the safe and precise operation of AI.
My genuine interest in Artificial Intelligence (AI) and established goal to advance AI safety and interpretability have driven me to apply for MSc in Artificial Intelligence. My extensive hands-on experience in autonomous driving technology, coupled with my academic research in Explainable AI, has underscored the necessity of understanding the underlying mechanisms of AI to ensure its safe and precise operation.
킹칼 교수
mechanistic interpretability
How does your experience and education make you a suitable candidate for this programme?
Starting with two years at the mobility startup Stryx, I began with connecting software and hardware, developing knowledge of sensor fusion, the starting point of the autonomous driving data pipeline. Later, I took responsibility for the latter part, vector map generation, providing overall insight into the data flow. This vertical understanding was instrumental in improving the pipeline, which established me as a key player in the company. My significant contributions led to the delivery of map for Korea’s first driverless service company, 42Dot, and facilitated Stryx's successful acquisition by Kakao Mobility. At Kakao Mobility, the company’s robo-taxi for public use offered firsthand experience that profoundly changed my attitude towards AI. After experiencing limitations, like stiffness, that industry experience alone could not resolve, I became convinced that a deeper academic understanding of AI was essential for making meaningful contributions to the field.
Driven by my resolve to delve deeper into the mechanisms of AI, my return to academia led me to focus on Explainable AI. Mentored by Professor Dongha Lee, I had the opportunity to delve into interpretable AI summarization. Our work in the RTSum (Relation-Triple Summarization) framework, which split and recombined information at a granular level, led to our paper’s acceptance at NAACL 2024. This not only enhanced my research skills but also confirmed my belief in the importance of a thorough understanding of the underlying computing for effectively directing LLMs. This blend of practical experience and academic research has prepared me to excel in the MSc in AI.
What do you hope to contribute to the computer science community, and how do you envision making a positive impact during your time in this programme? (max 250 words)
My belief in interpretable AI drives my scholarly engagement in the field. Inspired by Anthropic's research on Transformer models, I explored the roles of induction heads in in-context learning and conducted research on LLM functions using OpenAI's Transformer Debugger (TDB). I reinterpreted and shared this research with the AGI Korea group, highlighting my neuron analogy perspective.
At King's College London, I aim to deepen my exploration of Mechanistic Interpretability to enhance AI transparency and reliability. By collaborating with faculty and peers, I plan to develop innovative approaches to interpretable decision-making, building on my previous work in AI summarization and the RTSum framework.
I envision making a positive impact through active participation in research projects and scholarly publications. Sharing my insights at conferences and seminars will foster dialogue and collaboration, advancing the field. Additionally, I aim to mentor and support fellow students, creating an inclusive environment for idea exchange.
Applying my research to real-world challenges, particularly in LLM safety, is a key commitment. Leveraging the resources and expertise at King's College London, I am confident in my ability to advance AI safety and interpretability, making a lasting impact on the computer science community.