Geoffrey Hinton Quotes

In A.I., the holy grail was how do you generate internal representations.

When you have a large amount of data that is labeled so a computer knows what it means, and you have a large amount of computing power, and you're trying to find patterns in that data, we've found that deep learning is unbeatable.

A lot of what people are calling 'artificial intelligence' is really data analytics - in other words, business as usual. If the hype leaves you asking 'What is A.I., really?,' don't worry, you're not alone.

With the Google Brain project, we made the decision to build deep learning processes on top of Google's existing infrastructure.

People are going to use more and more AI. Acceleration is going to be the path forward for computing. These fundamental trends, I completely believe in them.

We really believe that long-term, the way AI will drive is similar to the way humans drive - we don't break the problem down into objects and vision and localization and planning. But how long it will take us to get there is questionable.

I get very excited when we discover a way of making neural networks better - and when that's closely related to how the brain works.

The challenge for what everybody's seeing in deep learning - the software richness is really quite high. In training, you have to wait days and weeks before it comes back to tell you whether your model works or not. And in the beginning, they all don't work.

Deep learning allows you to create predictive models at a level of quality and sophistication that was previously out of reach. And so deep learning also enhances the product function of data science because it can generate new product opportunities.

Everybody right now, they look at the current technology, and they think, 'OK, that's what artificial neural nets are.' And they don't realize how arbitrary it is. We just made it up! And there's no reason why we shouldn't make up something else.

Understanding vision and building visual systems is really understanding intelligence.

The paradigm for intelligence was logical reasoning, and the idea of what an internal representation would look like was it would be some kind of symbolic structure. That has completely changed with these big neural nets.

Now that neural nets work, industry and government have started calling neural nets AI. And the people in AI who spent all their life mocking neural nets and saying they'd never do anything are now happy to call them AI and try and get some of the money.

My main interest is in trying to find radically different kinds of neural nets.

The brain has about ten thousand parameters for every second of experience. We do not really have much experience about how systems like that work or how to make them be so good at finding structure in data.

I think the way we're doing computer vision is just wrong.

A deep-learning system doesn't have any explanatory power.

We have a good sense of the pulse of the industry. We know what's going on out there. We know that deep learning adoption is broad. It's going into production at scale.

If you step back a little and say we want to do A.I., then you will realize that A.I. needs knowledge, reasoning, and explanation.

I'd like to make a fundamental impact on one of the most exciting, intelligent questions of all time. Can we use software and hardware to build intelligence into a machine? Can that machine help us solve cancer? Can that machine help us solve climate change?

We now think of internal representation as great big vectors, and we do not think of logic as the paradigm for how to get things to work. We just think you can have these great big neural nets that learn, and so, instead of programming, you are just going to get them to learn everything.

The paradigm shift of the ImageNet thinking is that while a lot of people are paying attention to models, let's pay attention to data. Data will redefine how we think about models.

Computers are the most powerful tools that humanity has ever created. Yet, we treat them largely as a black box; as if it were an alien artifact that magically appeared on desks, in homes, and in our pockets.

Just like the brain consists of billions of highly connected neurons, a basic operating unit in a neural network is a neuron-like node. It takes input from other nodes and sends output to others.

Take any old classification problem where you have a lot of data, and it's going to be solved by deep learning. There's going to be thousands of applications of deep learning.