The 'Reward Scaling Pitfall' That 99% of Reinforcement Learning Engineers Fall Into

If you have ever written reinforcement learning code, you have almost certainly tried this. You scale the reward by a factor ...

Why Do Reasoning AIs Trained with Reinforcement Learning (RL) Lose to the Original Model After 100 Trials? Tsinghua University's pass@k Reversal and the METR 50-Mi…

Today, we are discussing the structural limitations of reasoning AIs and the duration of autonomous work, specifically why a ...
The Conversation

What is reinforcement learning? An AI researcher explains a key method of teaching machines – and how it relates to training your dog

Understanding intelligence and creating intelligent machines are grand scientific challenges of our times. The ability to learn from experience is a cornerstone of intelligence for machines and living ...
Forbes

The Rise And Rise Of Reinforcement Learning: AI’s Quiet Revolution

Forbes contributors publish independent expert analyses and insights. Author, Researcher and Speaker on Technology and Business Innovation. Apr 19, 2025, 03:24am EDT Apr 21, 2025, 10:40am EDT ...
techtarget

What is reinforcement learning?

Reinforcement learning (RL) is a machine learning training method that trains software to make certain desired actions. Reinforcement learning is based on rewarding desired behaviors and punishing ...
InfoWorld

3 ways to get into reinforcement learning

Whether you like theoretical study or want to get your hands dirty, plenty of reinforcement learning resources are out there. When I was in graduate school in the 1990s, one of my favorite classes was ...
BIOENGINEER

Contrastive Learning Tames Out-of-Distribution Actions in Offline Reinforcement Learning

Researchers at Hanyang University have developed TACCO, a contrastive learning method that explicitly identifies and ...
nsf.gov

Reinforcement Learning: The Engine Powering Today's AI Revolution

How do machines learn to walk, drive cars or chat like humans? These advances rely on a range of artificial intelligence techniques, including reinforcement learning, a branch of AI inspired by the ...
techtarget

What is reinforcement learning from human feedback (RLHF)?

Reinforcement learning from human feedback (RLHF) is a machine learning (ML) approach that combines reinforcement learning techniques, such as rewards and comparisons, with human guidance to train an ...