Glossary entry
आरएलएचएफ (मानव प्रतिक्रियाबाट सुदृढीकरण सिकाइ)
Definition (English)
Reinforcement Learning from Human Feedback — using human ratings to align model behavior.
अर्थ (नेपालीमा)
मानव प्रतिक्रियाबाट सुदृढीकरण सिकाइ — मोडेलको व्यवहार मिलाउन मानव मूल्याङ्कन प्रयोग गर्ने तरिका।
Related terms
Alignment
एलाइनमेन्ट / तालमेल
Ensuring an AI system's goals and behavior match human values and intentions.
Artificial Intelligence (AI)
कृत्रिम बुद्धिमत्ता
The field of building machines and software that can perform tasks normally requiring human intelligence, such as reasoning, learning, and decision-making.
Q-Learning
क्यू-लर्निङ
A reinforcement learning algorithm that learns the value of actions in each state.
Reinforcement Learning
सुदृढीकरण सिकाइ
Training an agent to make decisions by rewarding good actions and penalizing bad ones.
Active Learning
सक्रिय सिकाइ
A method where the model selects the most useful examples for humans to label.
Adam Optimizer
एडम अप्टिमाइजर
A widely used optimizer that adapts the learning rate for each parameter.
Affective Computing
भावनात्मक कम्प्युटिङ
AI that recognizes, interprets, or simulates human emotions.
Agentic AI
एजेन्टिक एआई
AI that autonomously plans and executes multi-step tasks using tools and feedback.