Day 5 of 180 | Reinforcement Model in Machine Learning | What is Reinforcement Learning day 5 Moye
10 Jan 2024 · 1 min

1:00
In this Hindi-language instructional video, Akansha Yadav explains how reinforcement learning (RL) underpins the learning mechanisms of ChatGPT. She describes RL as a feedback‑based learning approach where an agent receives rewards for positive actions and punishments for negative ones. When a user is unsatisfied with a response, they provide feedback, and through reinforcement learning from human feedback (RLHF) the model updates its behavior to produce more satisfactory, domain‑specific answers. The video emphasizes the role of rewards and punishments in shaping the model’s responses and illustrates how this process enables ChatGPT to improve over time based on user interactions.










