Will any transfer learning model, trained for any amount of time on one Atari environment, outperform the median human learning curve on most other Atari environments when transferred by 2026? | Manifold

Will any transfer learning model, trained for any amount of time on one Atari environment, outperform the median human learning curve on most other Atari environments when transferred by 2026?

Basic

7

Ṁ248

2026

45%

chance

1D

1W

1M

ALL

- Median human learning curve will be based on whatever the best available data for the human learning curve at the time is. If there is no good data available the question resolves N/A

This question is managed and resolved by Manifold.

#Technical AI Timelines

#Ancient Markets

Get

1,000

and

3.00

Sort by:

Any idea what the current performance is here?

@StephenMalina Essentially nonexistent. There are scattered attempts at transferring between a few games. AFAIK no one has made any notable progress and no one has tried transferring across all games.

Related questions

Will an AI achieve >85% performance on the FrontierMath benchmark before 2028?

Will an AI achieve >30% performance on the FrontierMath benchmark before 2026?

-15% 1d28% chance

In 2028, will an AI be able to play randomly selected computer games at human level without getting to practice?

Will an AI be able to play a type of video game that it wasn't trained on before 2026?

Will an AI model outperform 95% of Manifold users on accuracy before 2026?

Will any model get above human level (92%) on the Simple Bench benchmark before September 1st, 2025.

+22% 1d55% chance

Will a single model achieve superhuman performance on all Atari environments by 2025?

By 2026 will any RL agent with learned causal models of its environment achieve superhuman performance on >=10 Atari environments?

In 2028, will an AI be able to play randomly-selected computer games at human level, given the chance to train via self-play?

Will a new lab create a top-performing AI frontier model before 2028?

Related questions

Will an AI achieve >85% performance on the FrontierMath benchmark before 2028?

Will any model get above human level (92%) on the Simple Bench benchmark before September 1st, 2025.

Will an AI achieve >30% performance on the FrontierMath benchmark before 2026?

Will a single model achieve superhuman performance on all Atari environments by 2025?

In 2028, will an AI be able to play randomly selected computer games at human level without getting to practice?

By 2026 will any RL agent with learned causal models of its environment achieve superhuman performance on >=10 Atari environments?

Will an AI be able to play a type of video game that it wasn't trained on before 2026?

In 2028, will an AI be able to play randomly-selected computer games at human level, given the chance to train via self-play?

Will an AI model outperform 95% of Manifold users on accuracy before 2026?

Will a new lab create a top-performing AI frontier model before 2028?

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules