What is a reward function?
The scoring rule that tells a learning system how well it did, which shapes what it learns.
Ask your own in the app → TopicReinforcement learning · AI & Machine Learning
How does reinforcement learning train models?
Reinforcement learning trains a model by rewarding good outcomes. Instead of copying examples, the model tries things and gets a score. The papers we have covered use it to have a language model write the scoring rules for a level designer, and to retrain a mid-sized model to reason like a teacher.
Paper by paper
The papers we've covered
- AI & ML Reinforcement learning
ICLR 202010 min listen
Can Machines Invent Tools Nobody Taught Them?
Two teams of learning agents played hide-and-seek hundreds of millions of times with no instructions. They invented shelters, ramps, and a way to surf on boxes.
Peer-reviewed Simulation experiments1 environmentPeer-reviewed conference paper Moderate evidence
- AI & ML Reinforcement learning
arXiv8 min listen
Does Reward Training Make Chatbots Power-Hungry?
A classic worry says trained models learn to seek power. Tracked through training, one model's rate rose to 46.1%, then fell to 11.8%.
Preprint Checkpoint experiments1 model family Preliminary evidence
- AI & ML Reinforcement learning
Process Science7 min listen
Can a Small AI Learn to Draw Workflows That Run?
A code model broke most attempts to turn a written procedure into a working process map. Rewards a program can check cut failures to almost zero.
Peer-reviewed Training experiment with benchmark Moderate evidence
- AI & ML Reinforcement learning
Proceedings of the ACM on Software Engineering8 min listen
Can a Small AI Learn a Software Tester's Common Sense?
From a single example of a web service's reply, a small model trained with rewards writes the checks a tester would. Engineers kept over 96% of its tests.
Peer-reviewed Industrial evaluationOne company Limited evidence
- AI & ML Reinforcement learning
Frontiers in Artificial Intelligence8 min listen
Can a Mid-Sized AI Beat the Giants on a Teacher Exam?
A mid-sized open model, retrained to think like a teacher, scored 96.52% on a teacher exam benchmark. Catch: it trained on questions from the same benchmark.
Peer-reviewed Benchmark training experimentBrief report Limited evidence
- AI & ML Reinforcement learning
IEEE Transactions on Games8 min listen
Can an LLM Write the Rules That Train a Game Designer?
A language model writes the scoring rules that teach another AI to build game levels from a one-line story. With feedback, its rules matched hand-written ones.
Peer-reviewed Simulation experiments4 scenarios5 seeds Limited evidence
Questions people ask
Short answers, from the papers.
Can a language model write the rewards?
In one study, with a feedback loop, its rules matched hand-written ones on average.
Ask your own in the app →What is the risk?
A model can learn to chase the score itself, so results need checks on fresh, unseen tasks.
Ask your own in the app →Keep up with AI & Machine Learning research, ten minutes a paper.
The papers that matter in this hub, explained, and you can ask each one questions.
