DeepMind Achieves Holy Grail: An AI That Can Master Games Like Chess and Go Without Human Help

guest · Dec 6, 2018

DeepMind Achieves Holy Grail: An AI That Can Master Games Like Chess and Go Without Human Help
December 6, 2018
https://spectrum.ieee.org/tech-talk/robotics/artificial-intelligence/mb

DeepMind, the London-based subsidiary of Alphabet, has created a system that can quickly master any game in the class that includes chess, Go, and Shogi, and do so without human guidance.

The system, called AlphaZero, began its life last year by beating a DeepMind system that had been specialized just for Go.
The research, published today in the journal Science, was performed by a team led by DeepMind’s David Silver.

“This work has, in effect, closed a multi-decade chapter in AI research,” writes Campbell [...] “AI researchers need to look to a new generation of games to provide the next set of challenges.”
Click to expand...

guest · Feb 4, 2019

AI algorithm can play Montezuma's Revenge '10 times faster' than Google's DeepMind
February 4, 2019
https://www.theinquirer.net/inquire...revenge-10-times-faster-than-googles-deepmind

The reason why AI programmes struggle to master adventure games is that the 'rewards' in such games are randomly distributed.
Now, RMIT researchers claim their new algorithm can autonomously play Montezuma's Revenge and says it's 10 times quicker than DeepMind's AI in identifying subgoals, such as jumping over pits and climbing ladders.

The new algorithm learns from its earlier mistakes and has been developed by combining 'carrot-and-stick' reinforcement learning with an intrinsic motivation approach, which offers rewards to the programme for being curious and autonomously exploring useful sub-goals.

"We've shown that the right kind of algorithms can improve results using a smarter approach rather than purely brute forcing a problem end-to-end on very powerful computers," says Professor Zambetta.
Professor Zambetta also believes their work could have potential application outside of gaming, for example, to improve algorithms controlling self-driving cars or robotic assistants, which are required to achieve goals in the real world.
Click to expand...

guest · Feb 24, 2019

How AlphaZero has rewritten the rules of game play on its own
February 22, 2019
https://www.technologyreview.com/s/...s-rewritten-the-rules-of-gameplay-on-its-own/

Silver’s latest creation, AlphaZero, learns to play board games including Go, chess, and Shogi by practicing against itself. Through millions of practice games, AlphaZero discovers strategies that it took humans millennia to develop.
So could AI one day solve problems that human minds never could? I spoke to Silver at his London office at DeepMind, now owned by Alphabet.
Since AlphaZero doesn’t learn from humans, is it even more creative?
AlphaZero has to figure out everything for itself. Every single step is a creative leap. Those insights are creative because they weren’t given to it by humans. And those leaps continue until it is something that is beyond our abilities and has the potential to amaze us.
You’ve had AlphaZero play against the top conventional chess engine, Stockfish. What have you learned?
AlphaZero learns to understand positions for itself.
It’s found a way to understand the position which is unthinkable according to the norms of chess. It understands it’s better to have the position than the four pawns.
Click to expand...

If we think about the capabilities of the human mind, we’re still a long way away from achieving that.
Click to expand...

guest · Jan 17, 2020

AlphaZero beat humans at Chess and StarCraft, now it’s working with quantum computers
January 16, 2020
https://thenextweb.com/artificial-i...craft-now-its-working-with-quantum-computers/

A team of researchers from Aarhus University in Denmark let DeepMind‘s AlphaZero algorithm loose on a few quantum computing optimization problems and, much to everyone’s surprise, the AI was able to solve the problems without any outside expert knowledge. Not bad for a machine learning paradigm designed to win at games like Chess and StarCraft.

When the Aarhus team applied AlphaZero’s optimization abilities to a trio of problems associated with optimizing quantum functions – an open problem for the quantum computing world – they learned that its ability to learn new parameters unsupervised transferred over from games to applications quite well.
Per the study: [...]

Before we start getting too concerned about unsupervised AI accessing quantum computers, it’s worth mentioning that so far AlphaZero’s just solved a few problems in order to prove a concept. We know the algorithms can handle quantum optimization, now it’s time to figure out what we can do with it.
Click to expand...

guest · Feb 23, 2020

Defeated Chess Champ Garry Kasparov Has Made Peace With AI
Twenty-three years after he lost to Deep Blue, Kasparov says people need to work with machines
February 21, 2020
https://www.wired.com/story/defeated-chess-champ-garry-kasparov-made-peace-ai/

Garry Kasparov is perhaps the greatest chess player in history. For almost two decades after becoming world champion in 1985, he dominated the game with a ferocious style of play and an equally ferocious swagger.
In 1997, at the height of his powers, Kasparov was crushed and cowed by an IBM supercomputer called Deep Blue.

More significantly, thanks to recent progress in artificial intelligence, machines are learning and exploring the game for themselves.
Deep Blue followed hand-coded rules for playing chess. By contrast, AlphaZero, a program revealed by the Alphabet subsidiary DeepMind in 2017, taught itself to play the game at a grandmaster level simply by practicing over and over. Most remarkably, AlphaZero uncovered new approaches to the game that dazzled chess experts.
Click to expand...

Last week, Kasparov returned to the scene of his famous Deep Blue defeat—the ballroom of a New York hotel—for a debate with AI experts organized by the Association for the Advancement of Artificial Intelligence. He met with WIRED senior writer Will Knight there to discuss chess, AI, and a strategy for staying a step ahead of machines
Click to expand...

guest · Apr 1, 2020

DeepMind’s Agent57 beats humans at 57 classic Atari games
March 31, 2020
https://venturebeat.com/2020/03/31/deepminds-agent57-beats-humans-at-57-classic-atari-games/

In a preprint paper published this week by DeepMind, Google parent company Alphabet’s U.K.-based research division, a team of scientists describe Agent57, which they say is the first system that outperforms humans on all 57 Atari games in the Arcade Learning Environment data set.

Assuming the claim holds water, Agent57 could lay the groundwork for more capable AI decision-making models than have been previously released.

“With Agent57, we have succeeded in building a more generally intelligent agent that has above-human performance on all tasks in the Atari57 benchmark,” wrote the study’s coauthors. “Agent57 was able to scale with increasing amounts of computation: the longer it trained, the higher its score got.”
To evaluate Agent57, the DeepMind team compared it with leading algorithms including MuZero, R2D2, and NGU alone. [...] Agent57 showed greater capped mean performance (100) versus both R2D2 (96.93) and MuZero (89.92), taking 5 billion frames to surpass human performance on 51 games and 78 billion frames to surpass it in Skiing.

“Agent57 finally obtains above human-level performance on the very hardest games in the benchmark set, as well as the easiest ones,” wrote the coauthors in a blog post.
Click to expand...

DeepMind: Agent57: Outperforming the human Atari benchmark

The Atari57 suite of games is a long-standing benchmark to gauge agent performance across a wide range of tasks. We’ve developed Agent57, the first deep reinforcement learning agent to obtain a score that is above the human baseline on all 57 Atari 2600 games. Agent57 combines an algorithm for efficient exploration with a meta-controller that adapts the exploration and long vs. short-term behaviour of the agent.

Read the paper here.
Click to expand...

guest · Dec 23, 2020

DeepMind's latest AI can master games without being told their rules
MuZero represents a likely breakthrough in general-purpose AI
December 23, 2020
https://www.engadget.com/deepmind-muzero-160024950.html

In 2016, Alphabet's DeepMind came out with AlphaGo, an AI which consistently beat the best human Go players. One year later, the subsidiary went on to refine its work, creating AlphaGo Zero.
Where its predecessor learned to play Go by observing amateur and professional matches, AlphaGo Zero mastered the ancient game by simply playing against itself.

DeepMind's latest AI, MuZero, didn't need to be told the rules of go, chess, shogi and a suite of Atari games to master them. Instead, it learned them all on its own and is just as capable or better at them than any of DeepMind's previous algorithms.
Click to expand...

It will consider the outcome of its previous decision, the current position it finds itself in and the best course of action to take next. That seemingly simple approach makes MuZero the most effective algorithm DeepMind made to date.
In its testing, it found MuZero was as good as AlphaZero at chess, Go and shogi, and better than all its previous algorithms, including Agent57, at Atari games. It also found that the more time it gave MuZero to consider an action, the better it performed.
Click to expand...

Creating an algorithm that can adapt to a situation where it doesn't know all the rules governing a simulation, but it can still find a way to plan for success has been a challenge AI researchers have been trying to solve for a while.
Click to expand...

Log in or Sign up

DeepMind Achieves Holy Grail: An AI That Can Master Games Like Chess and Go Without Human Help

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

Log in or Sign up

DeepMind Achieves Holy Grail: An AI That Can Master Games Like Chess and Go Without Human Help

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

guest Guest

Useful Searches