How close are we to building a truly intelligent agent?
Most scientists think we are still decades away, but today, a group of scientists from @DeepMind claims they know how to get there.
Let's talk about what's going on.
Most scientists think we are still decades away, but today, a group of scientists from @DeepMind claims they know how to get there.
Let's talk about what's going on.
What is "Artificial General Intelligence" (AGI)?
An agent capable of learning any intellectual task than a person can also learn.
Today, AI has been limited to systems that can learn particular tasks. A system that can learn anything you teach it, just like a human, is AGI.
An agent capable of learning any intellectual task than a person can also learn.
Today, AI has been limited to systems that can learn particular tasks. A system that can learn anything you teach it, just like a human, is AGI.
Unfortunately, there's no way to build such a general, intelligent agent without formulating a custom solution for every individual task.
This sucks. This doesn't scale. This doesn't get us to AGI.
But maybe we aren't that far off...
This sucks. This doesn't scale. This doesn't get us to AGI.
But maybe we aren't that far off...
"Reward Is Enough" is a paper from DeepMind, the company behind AlphaGo and AlphaZero.
Their claim:
"Reinforcement learning agents could constitute a solution to artificial general intelligence."
Their claim:
"Reinforcement learning agents could constitute a solution to artificial general intelligence."
Reinforcement Learning is a well-studied branch of Machine Learning that's based on reward maximization and trial-and-error experience acquisition.
The paper suggests that these characteristics are enough to build agents that exhibit intelligence.
The paper suggests that these characteristics are enough to build agents that exhibit intelligence.
How we've been tackling problems:
Break it down, build components that solve each piece, and connect them with some logical glue.
But this is not how natural intelligence works.
Break it down, build components that solve each piece, and connect them with some logical glue.
But this is not how natural intelligence works.
DeepMind's hypothesis:
"(...) the generic objective of maximizing reward is enough to drive behavior that exhibits most if not all abilities that are studied in natural and artificial intelligence."
In other words: We can get to AGI by mimicking how nature works.
"(...) the generic objective of maximizing reward is enough to drive behavior that exhibits most if not all abilities that are studied in natural and artificial intelligence."
In other words: We can get to AGI by mimicking how nature works.
We have been studying Reinforcement Learning for quite a long time, so we know it well, and we have made impressive progress using it.
This makes me hopeful. We might reach AGI sooner than we thought!
This makes me hopeful. We might reach AGI sooner than we thought!
Of course, there are still many challenges we need to solve.
Reinforcement learning needs a lot of data to gain experience. Designing a reward system is hard, and we still don't know how to build systems that work across different domains.
More work needs to happen.
Reinforcement learning needs a lot of data to gain experience. Designing a reward system is hard, and we still don't know how to build systems that work across different domains.
More work needs to happen.
Let me give you a few interesting links.
First, here is the full paper, in case you want to read it:
sciencedirect.com/science/articl…
First, here is the full paper, in case you want to read it:
sciencedirect.com/science/articl…
David Silver is one of the authors of the paper.
A couple of years ago, I took his course on Reinforcement Learning.
It's 100% free and published on YouTube.
Here's the playlist with almost 20 hours of materials: youtube.com/playlist?list=…
A couple of years ago, I took his course on Reinforcement Learning.
It's 100% free and published on YouTube.
Here's the playlist with almost 20 hours of materials: youtube.com/playlist?list=…
Last year, David was in the Lex Fridman podcast talking about the incredible systems he has helped build: .
Doina Precup, another of the authors of the paper, gave a TED talk on how computers learn like people:
Doina Precup, another of the authors of the paper, gave a TED talk on how computers learn like people:
Finally, here is a full documentary on Alpha Go.
An incredibly compelling story on how the system came to be and achieved what nobody thought possible before.
Watch it, and it's going to be the highlight of your week.
An incredibly compelling story on how the system came to be and achieved what nobody thought possible before.
Watch it, and it's going to be the highlight of your week.
I post threads like this every week.
Stay tuned as I help you get to the core of practical machine learning.
You can find the rest of my threads here: @svpino.
Stay tuned as I help you get to the core of practical machine learning.
You can find the rest of my threads here: @svpino.
I got a lot from it. Maybe because I already had a background in Reinforcement Learning.
https://twitter.com/_amirrahnama/status/1404379447185248258
Absolutely yes.
https://twitter.com/ehsgngr8/status/1404412322010042368
• • •
Missing some Tweet in this thread? You can try to
force a refresh
