AI & ML Glossary

A

• Ablation study (19)
• Agent (60)
• Alignment (60)
• Attention mechanism (62)

B

• Benchmark (60)

C

• Chain of thought (29)
• Context window (6)
• Convolutional neural network (4)
• Cross-attention (15)
• Cross-modal (2)
• Curriculum learning (2)

D

• Diffusion model (15)
• Distillation (48)

E

• Embeddings (37)
• Embodied AI
• Emergent (3)
• Ensemble

F

• Fine-tuning (144)
• Flow matching (15)

G

• Generalization (5)
• Ground truth
• Grounding (16)
• Group relative policy optimization (35)
• Guardrails

H

• Hallucination (20)
• Held-out set
• Hyperparameter (3)

I

• Image segmentation (2)
• Imitation learning (2)
• In-context learning (21)
• Inference (60)

J

• Jailbreak (4)

K

• KV cache (21)

L

• Large language model (172)
• Latent space (12)
• LoRA (28)

M

• Mixture of experts (24)
• Multi-agent system (4)
• Multimodal (60)

N

• Neural network (15)

O

• On-policy distillation (17)
• Orchestration (3)
• Out-of-distribution (5)
• Overfitting (3)

P

• Parameters (9)
• Pass@k
• Policy (46)
• Policy gradient (2)
• Policy network
• Power law (3)
• PPO (2)
• Precision (3)
• Pretraining (14)
• Prompt injection (8)

Q

• Quantization (22)

R

• Recall (5)
• Red teaming
• Reinforcement learning (71)
• Retrieval-augmented generation (35)
• Reward hacking (7)
• Reward model (5)
• Reward signal
• RLHF (6)
• Router (15)

S

• Self-consistency (4)
• Self-distillation (17)
• Sim-to-real
• Supervised fine-tuning (34)
• Supervised learning

T

• Test-time scaling (9)
• Token (38)
• Tool use (7)
• Trajectory (38)
• Transformer (76)
• Tree of thoughts

V

• Vision transformer (5)
• Vision-language model (55)
• Vision-language-action model (15)

W

• World model (19)

Z

• Zero-shot (27)