6 takeaways from Professor Xiong's lecture in GUi

Professor Xiong delivers a keynote speech about algorithms. [Photo/GUi]

Xiong Hui, chair professor and associate vice president of the Hong Kong University of Science and Technology (Guangzhou), also shared several standpoints in his keynote address in Grand Union of Innovation (GUi). Here are the takeaways: 

1. The previous generation of technological advancements mainly reduced manual labor, whereas the progress in large language models (LLM) nowadays saves toilsome routines of intellectual labor.

2. The genuine industrial value is lost in your daily engagement of tasks that start from the scratch instead of continuous accumulation. 

3. What we need are more innovative minds rather than "golden fingers" in domestic industry. The former entails innovations from the ground, while the latter involves leaps from 1 to N.

4. Algorithms should also strive for "moderation," aiming for a balance between bias and variance.

5. Humans possess two gears of learning, either fast or slow. Whether individuals or organizations excel depends on their ability in slow learning.

6. We are ourselves a process of model training, constantly refining our models and parameters from childhood to adulthood so as to facilitate our careers.