Sharing — Fei-Fei Li: Artificial Intelligence is on its way to reshape the world

Fei-Fei Li, Who Builds Large-Scale Dataset, ImageNet, for AI

3 min readOct 19, 2022

Today, I would like to share about a Q&A Interview of Fei-Fei Li, which is also an article of National Science Review (NSR), in 2017:

In 2017, she was on sabbatical, worked as Chief Scientist of AI/ML of Google Cloud, until the second half of 2018.
“It is time for AI to help other vertical industries like healthcare, agriculture and manufacture to transform and upgrade. Google Cloud is an excellent platform that will accelerate this process, which has both scientific and commercial significance.” — Li.

“There is close cooperation of AI scientists and neural scientists is an emerging phenomenon in China.” responded by NSR Interviewer .
“I think the most practical way for AI to leap forward is to break the bottleneck of lacking interpretability, architectural knowledge and training flexibility of neural networks.” — Li.

“I suspect that the propaganda of AGI is motivated for commercial interest by people who have no idea about what it really means. … If we need to define AGI, maybe it is an agent capable of multi-knowledge presentation, multi-sensory, multi-layer reasoning and learning. … Cars run faster than us; cranes hold heavier objects; it is not necessary to be afraid (of AGI).” — Li.

“When I began to build ImageNet, there was little funding for me doing this. But I did not give up.” — Li

“Based on statistics, there is no way that would guarantee success, and it is especially true in research.” — She Continued.

“‘Quantity Breeds Quality’, incremental research is also important. The Deep Residual Networks (ResNets) proposed by Kaiming He, currently the prevalent architecture in computer vision, has shown great potential in areas like NLP, speech recognition. I hope there will be more and more ‘Kaiming He’ in China.” — Li.

“AI is a ‘real deal’, … I firmly believe AI is as ‘real’ as computing, the Internet, renewable energy, new materials, etc.” — Li.

“The truth is, at this moment, this is the time to double down on basic, fundamental research in AI. It’s time to support and fund long-term research that address many of the most difficult and yet to be solved problems of AI.” — Li.
“For NLP, we still stay at a relatively superficial level, lacking deep interactive dialogue. For CV, although products based on it are emerging, we are far from fulfilling our aim.” — She mentioned some examples.

At the end, I would like to share 2 of her talks.
One is the TED Talk in 2015, about constructing the ImageNet dataset for AI, as below:

TedTalk about Constructing ImageNet for AI in 2015

This TED Talk, talking about ImageNet for AI, has already reached over 2M views in TED, and over 1M views in YouTube. The corresponding papers in 2009 CVPR and 2015 IJCV have gained over 40000 and 30000 citations respectively.
Another one is the AnitaB.org talk in 2018:

AI & ML Talk at AnitaB_org in 2018

Besides talking about ImageNet, she also mentioned about image captioning research, as well as AI for satellite imaging, under water imaging, and sensors at home, etc.

2017 [AI Reshapes World] 2019 [New Heights with ANN] 2021 [Deep Learning for AI] 2022 [Small is the New Big]