Go to Zero
Вставка
- Опубліковано 3 жов 2024
- Selected as “New and Noteworthy” by Apple Podcasts, the highly-praised, award-nominated first season of "DeepMind: The Podcast" explores the fascinating world of artificial intelligence (AI). Join mathematician and broadcaster Hannah Fry as she meets world-class scientists and thinkers as they explain the foundations of AI, explore some of the challenges the field is wrestling with, and dives into the research that's led to breakthroughs like AlphaGo and AlphaFold. Whether you’re a beginner or an experienced researcher, join our journey into the past, present, and future of AI.
In episode 2, Hannah meets the team behind AlphaGo. In 2016, more than 200 million people watched it become the first computer program to defeat a professional human player at the game of Go, a milestone in AI research. Since then, the DeepMind team has continued to develop the system - including AlphaZero, a program that taught itself how to play chess, Go, and shogi.
Joined in this episode by Chess Grandmaster Matthew Sadler and Women’s International Master Natasha Regan (co-authors of Game Changer: AlphaZero's groundbreaking chess strategies and the promise of AI), DeepMind researchers tell the story of AlphaZero and explain why games continue to be such useful proving ground for AI research.
“Creativity takes us out of our expected patterns of behaviour. In that sense, the program was truly creative.” - David Silver, Lead Researcher at DeepMind
#AlphaGo #AlphaZero #worldchampion #creativity #artificialintelligence #DMpodcast
_ _
Listen to the full series on UA-cam: dpmd.ai/3geDPmL
Or search for “DeepMind: The Podcast” on your favourite podcast app, including:
Search “DeepMind: The Podcast” and subscribe on your favourite podcast app.
Apple Podcasts: dpmd.ai/2Rzlmcu
Google Podcasts: dpmd.ai/3geDjp5
Spotify: dpmd.ai/3w29cb4
Pocket Casts: pca.st/30m1
"It turns out we have a tendency as human designers to believe that we know how to make a system stronger. But quite often it turns out that by putting our own predispositions and preferences into our programs, we actually make them weaker." :)
"AlphaGo - The Movie" is a true piece of art that tells this story. It can be found here in UA-cam, it's not very technical but on the human side it's amazing.
4:36 "It's played on a pale wooden board engraved with an elegant grid of 19 by 19 squares".
Slip of the tongue boundary error. Should be 18 by 18 squares (giving 19 by 19 intersections).
If you ever consider changing it to a videocast would be so helpful. Seeing people and emotions would be awesome. Thanks!!
The triumph of AlphaGo over humans is what really changed my mind of how close we're to AGI.
The approach to AI learning here seems to be primarily rewards based, but I recall that HAL in the Kubrick film. "2001", achieved its learning through fear, jealousy, and other sloppy motivators. HAL was paranoid. Artificial Intelligence needs to be exposed to the full range of human emotion, motivation, and dismotivation before it can truly reach singularity.
But of course, that prospect of singularity is exactly what makes humans fearful and paranoid of Artificial Intelligence!
13:40 Silver:
I think we need to challenge ourselves to ask "What is creativity?"
I think creativity should be defined as anything which takes us out of our expected patterns of behavior.
In addition to Silver's words, I would say that creativity must cause good results, victory in the case of go. Amateur players make bold moves, like move 37, much more frequently than pros, but did not convert those random moves, so to speak, into victory. The beauty of move 37 is that it actually seemed to work best.
Love the podcast! Interesting narrative, dialogs and sound arrangement. What’s the piano piece playing at 19:25?
This is brilliant. Thank you for the publication!
In reality David whispered to AlphaGo in the backstage "dude lose just this one and we'll share the profit, mkay?"
i hope there are more of these
This is a great episode!
When I reached the 23:23,I realized that we can deploy our political system against the democracy system in reinforcement learning,to see the republic of Plato
So great!
Where are the show notes?
David Silver is a really cool cat.
I love her voice
Computers have nothing holding them back, we can learn so much from how a computer would respond to a situation. I can't wait until computers like this are the norm ^_^. It's so amazing, our brains are so basic compared to what a computer can get to.
Yours is.
Actually our brains are very complex and hard to understand, but work similarly to those neural networks (the "brain" behind those computers). This technology was invented by basically copying neurons artificially.
Human neural networks are far superior. It's our communication bandwidth that limits us. AlphaGo played horribly studying thousands of games (human masters and/or its own). Only after trillions of games does it achieve superhuman strength. Humans can't even play a million games only because of I/O limits. It's not memory either. Neither must remember past games.
grat to see alphazero rock but why didnt deepmind have thing like chatgpt
hey deepmind when you are making AlphaCOD?
yeet