Glossary · letter R
AI terms starting with R.
31 entries that start with R, each with its Nepali equivalent and a plain-language definition. Click any term to open its dedicated page.
-
R-squared
आर-वर्ग
A measure of how much of the variation in data a regression model explains.
-
Random Forest
र्यान्डम फरेस्ट
An ensemble of many decision trees whose predictions are combined for better accuracy.
-
Range
परास / दायरा
The difference between the largest and smallest values in a dataset.
-
Rate Limit
रेट लिमिट / दर सीमा
A cap on how many requests a user can make to a service in a time window.
-
Rational Agent
तर्कसंगत एजेन्ट
An entity that perceives its environment and acts to achieve the best expected outcome.
-
Reasoning
तर्क / विवेचना
Drawing conclusions from facts and rules to solve problems or make decisions.
-
Recall
रिकल / पुनःप्राप्ति
Of all actual positive items, the fraction the model correctly identified.
-
Recommendation System
सिफारिस प्रणाली
An AI system that suggests items to users based on their behavior and preferences.
-
Recurrent Neural Network (RNN)
रिकरेन्ट न्युरल नेटवर्क
A network designed for sequences that remembers previous inputs as it processes data.
-
Red Teaming
रेड टिमिङ
Deliberately probing an AI system for weaknesses and harmful behaviors before release.
-
Regression
रिग्रेसन / प्रतिगमन
A task where the model predicts a continuous numeric value, such as price or temperature.
-
Regularization
रेगुलराइजेसन
Techniques that penalize complexity to prevent overfitting and improve generalization.
-
Regulation
नियमन
Laws and official rules that govern how AI may be built and used.
-
Reinforcement Learning
सुदृढीकरण सिकाइ
Training an agent to make decisions by rewarding good actions and penalizing bad ones.
-
ReLU
रेलु (ReLU)
A popular activation function that outputs the input if positive and zero otherwise.
-
Repetition Penalty
पुनरावृत्ति दण्ड
A setting that discourages a model from repeating the same words or phrases.
-
Reproducibility
पुनरुत्पादनीयता
The ability to obtain the same results by re-running a model or experiment.
-
Reranking
रिर्याङ्किङ
Reordering retrieved results so the most relevant ones appear first.
-
Residual Network (ResNet)
रेसिड्युअल नेटवर्क
A deep network design using skip connections to train very deep models effectively.
-
Responsible AI
जिम्मेवार एआई
An approach to designing AI that prioritizes ethics, safety, and societal benefit.
-
REST API
रेस्ट एपीआई
A common style of web API for sending and receiving data over standard requests.
-
Retrieval-Augmented Generation (RAG)
रिट्रिभल-अगमेन्टेड जेनेरेसन
Combining a language model with a search system so answers are grounded in retrieved documents.
-
Return on Investment (ROI)
लगानी प्रतिफल
A measure of the financial benefit gained relative to the cost of an AI project.
-
Reward Function
पुरस्कार फंक्सन
The rule that defines what outcomes an RL agent is rewarded for achieving.
-
RGB
आरजीबी (रातो-हरियो-नीलो)
A color model that represents colors as mixtures of red, green, and blue.
-
Right to Explanation
व्याख्याको अधिकार
The idea that people affected by an automated decision deserve to know its reasoning.
-
Risk Assessment
जोखिम मूल्याङ्कन
Identifying and evaluating the potential harms of deploying an AI system.
-
RLHF
आरएलएचएफ (मानव प्रतिक्रियाबाट सुदृढीकरण सिकाइ)
Reinforcement Learning from Human Feedback — using human ratings to align model behavior.
-
Robotic Process Automation (RPA)
रोबोटिक प्रक्रिया स्वचालन
Software bots that automate repetitive, rule-based digital tasks in business.
-
Robotics
रोबोटिक्स
The field of designing and building machines that sense and act in the physical world.
-
ROC Curve
आरओसी कर्भ
A graph showing the tradeoff between true positive and false positive rates.