BUILD YOUR AI VOCABULARY

AI Vocabulary

225+terms

Artificial Intelligence (AI)Machine Learning (ML)Deep Learning (DL)Data Input Output Function (Rule-based Logic)Pattern Automation

Dataset Feature Label Algorithm Model Training Prediction Inference

Training Data Validation Data Test Data Train-Test Split Ground Truth Data Preprocessing Normalization Feature Engineering Annotation Labeling Data Augmentation

Parameters Weights Bias (model parameter)Hyperparameters Loss Function Gradient Gradient Descent Learning Rate Batch Batch Size Epoch Backpropagation Optimization Checkpoint Seed

Generalization Overfitting Underfitting Regularization Dropout Cross-Validation Validation

Baseline Benchmark Evaluation Accuracy Confusion Matrix Precision Recall F1 Score

Neural Network Node Layer Input Layer Hidden Layer Output Layer Activation Function ReLU

Token Tokenization Sequence Vector Embedding Embedding Model Dense Vector Sparse Vector Vector Database Vector Search Approximate Nearest Neighbor (ANN)HNSW Cosine Similarity Euclidean Distance Dot Product Similarity Search Semantic Search

Language Model Large Language Model (LLM)Foundation Model Generative AI Next-Token Prediction Logits Probability Distribution Sampling

Transformer Encoder Decoder Attention Self-Attention Positional Encoding Context Window Context Length KV Cache

Prompt System Prompt Prompt Engineering Zero-Shot Few-Shot Chain-of-Thought Temperature Top-k Top-p Reasoning Test-Time Compute Inference-Time Scaling Thinking Tokens

Pretraining Transfer Learning Fine-Tuning Instruction Tuning Parameter-Efficient Fine-Tuning (PEFT)LoRA QLoRA Self-Supervised Learning

Reinforcement Learning (RL)RLHF DPO Reward Model Preference Data Alignment AI Safety Guardrails

Mixture of Experts (MoE)Sparse Model Dense Model Parameter Count Model Architecture Scaling Laws Emergent Abilities Distillation Pruning

GAN (Generative Adversarial Network)Autoencoder Latent Space Diffusion Model VAE Text-to-Image Text-to-Video Image-to-Image Multimodal Generation Generative Model

Multimodal AI Multimodal Model Vision-Language Model (VLM)Vision Encoder Computer Vision Image Generation OCR Speech-to-Text Text-to-Speech

Explainability Interpretability AI Bias Prompt Injection Jailbreak Adversarial Attack Data Poisoning Data Leakage Model Poisoning Red Teaming Responsible AI

GPU TPU CUDA VRAM Inference Server Model Serving Batch Inference Streaming Latency Throughput Tokens Per Second (TPS)Quantization FP32 FP16 BF16 INT8 INT4

LLM Evaluation Human Evaluation Automated Evaluation LLM-as-a-Judge Evaluation Dataset Faithfulness Answer Correctness Toxicity Relevance Context Precision Context Recall

Agentic AI / Autonomous Agents Orchestration / Tool-Calling Open-Source vs Closed-Source Models Tracing Logging Scalability Cost per Token Token Usage