Artificial Intelligence
35 views
AI Alignment
Quick Definition
Ensuring AI systems act in accordance with human values
Full Definition
Ensuring AI systems behave in accordance with human values, intentions, and goals.
Examples
value learning, reward modeling, constitutional AI
Related Terms
ai-safety
ai-ethics
More Artificial Intelligence Terms
Multimodal AI
AI that processes and integrates multiple data types
Quantization
Reducing model precision from 32-bit to lower bit representations
Prompt Engineering
Crafting effective inputs to guide AI model outputs
AI Agent
Autonomous software taking goal-directed actions
Reward Modeling
Training models to predict human preferences for RLHF
Meta-Learning
Designing models that learn how to learn new tasks