ReasoningGym: Reasoning Environments for RL with Verifiable Rewards

Re:ReasoningGym: Reasoning Environments for RL with Verifiab... by flowerthoughts on Monday June 02, 2025 @07:15AM (#44157610)

Assuming with GDM, you mean Google-Deep Mind. They pioneered RL with deep nets as policy function estimator. The deep nets being a result of CNNs and massive improvements in hardware parallelization at the time.

RL was established, at the latest, with Q-learning in 1989: https://en.wikipedia.org/wiki/Q-learning

[ Parent | on HN ]

Re:ReasoningGym: Reasoning Environments for RL with Verifiab... by t55 on Monday June 02, 2025 @10:08AM (#44159025)

i didn't say they invented everything; in science you always stand on the shoulders of giants

i still think my original statement is fair

[ Parent | on HN ]

Re:ReasoningGym: Reasoning Environments for RL with Verifiab... by lechatonnoir on Monday June 02, 2025 @04:51PM (#44162922)

"gdm pioneered rl" is definitely not actually right, but it's correct to assert that they were huge players.

people who knew from context that your statement was broadly not actually right would know what you mean and agree on vibes. people who didn't could reasonably be misled, i think.

[ Parent | on HN ]