Name: These virtual robot arms get smarter by training each other
Brand: technologyreview.com
SKU: technologyreview-com-127
Availability: OnlineOnly

critic

These virtual robot arms get smarter by training each other

preview

You need to enable JavaScript to view this site.A virtual —stacking blocks, setting the table, arranging chess pieces—without having to be retrained for each task. It did this by playing against a second robot arm that was trained to give it harder and harder challenges. Self play: Developed by researchers at , the identical robot arms—Alice and Bob—learn by playing a game against each other in a simulation, without human input. The robots use reinforcement learning, a technique in which AIs are trained by trial and error what actions to take in different situations to achieve certain goals. The game involves moving objects around on a virtual tabletop. By arranging objects in specific ways, Alice tries to set puzzles that are hard for Bob to solve. Bob tries to solve Alice’s puzzles. As they learn, Alice sets more complex puzzles and Bob gets better at solving them. Multitasking: Deep-learning models typically have to be retrained between tasks. For example, AlphaZero (which also learns by playing games against itself) uses a single algorithm to teach itself to play chess, shogi and Go—but only one game at a time. The chess-playing AlphaZero cannot play Go and the Go-playing one cannot play shogi. Building machines that really can multitask is a big unsolved problem on .&nbsp;&nbsp; AI dojo:&nbsp; One issue is that training an AI to multitask requires a vast number of examples. OpenAI avoids this by training Alice to generate the examples for Bob, using one AI to train another. Alice learned to set goals such as building a tower of blocks, then picking up it up and balancing it. Bob learned to use properties of the (virtual) environment, such as friction, to grasp and rotate objects. Virtual reality: So far the approach has only been tested in a simulation but researchers at OpenAI and elsewhere are getting better at transferring models trained in virtual environments to physical ones. A simulation lets AIs churn through large datasets in a short amount of time, before being fine-tuned for real-world settings.&nbsp;&nbsp; Overall ambition: The researchers say that their ultimate aim is to train a robot to solve any task that a person might ask it to. Like GPT-3, a language model that can use language in a wide variety of different ways, these robot arms are part of OpenAI’s overall ambition to build a multitasking AI. Using one AI to train another could be a key part of that. Large language models are trained on troves of personal data hoovered from the internet. So I wanted to know: What does it have on me?What Gran Turismo Sophy learned on the racetrack could help shape the future of machines that can work alongside humans, or join us on the roads.And it’s giving the data away for free, which could spur new scientific discoveries.But the company has had to rush out fixes to the image-making model’s worst flaws to do so.Discover special offers, top stories,upcoming events, and more.

You need to enable JavaScript to view this site.A virtual —stacking blocks, setting the table, arranging chess pieces—without having to be retrained for each task. It did this by playing against a second robot arm that was trained to give it harder and harder challenges. Self play: Developed by researchers at , the identical robot arms—Alice and Bob—learn by playing a game against each other in a simulation, without human input. The robots use reinforcement learning, a technique in which AIs are trained by trial and error what actions to take in different situations to achieve certain...

public

Author Will Douglas Heaven

A media company making technology a greater force for good. Get our journalism: http://technologyreview.com/newsletters

technologyreview.com

Credible

Speculation

Anecdotal Evidence

Generalization

Stacking the Deck

Appeal to Authority

Straw Man

Slippery Slope

Emotional Appeal

Red Herring

False Dilemma

Illogical

Moral Equivalency

Ad Hominem

Correlation w/o Causation

Semantics

Begging the Question

Faulty Analogy

Non Sequitur

Bandwagon

Dogmatism

Failing Occam's Razor

Biased

Pure Opinion

Hit Piece

Racial Bias

National Bias

Religious Bias

Financial Incentive

Political Agenda

Gender Bias

Misused Image/Video

Study Misinterpreted

Mistake

Misused Term

Science Misrepresented

Factual Error

Satire

Surface Level

Lack of Reliable Sources

Sensational

Not Credible

Investigative

Balanced

Well Sourced

Great Context

sports

entertainment

U.S.

international

business

technology

science

culture

politics

media

health

These virtual robot arms get smarter by training each other

No article scores yet.

CRITIC REVIEWS

PUBLIC REVIEWS