Reinforcement learning

AI fundamentals

Reinforcement learning improves an agent through trial and error: it acts in an environment, receives rewards or penalties, and adjusts its strategy to maximise the gain. In the form of RLHF (reinforcement learning from human feedback), it is also a key step in aligning large language models.

In practice at Gensai

Gensai follows these techniques closely: they explain much of the behaviour of the models the agency integrates.

Same theme

Artificial intelligence (AI)Algorithm Data Dataset Model (AI)Machine learning Deep learning Neural network Supervised learning Unsupervised learning Classification Regression Prediction Training Inference Parameters Hyperparameters Bias (AI)