Flashcards › Technology & Coding › Artificial Intelligence
AI and Machine Learning Terms
Key ideas in artificial intelligence, from training data to neural networks.
0 of 71 known
71 cards
- Artificial intelligence (AI)TermComputer systems that do tasks that usually need human intelligence
- Machine learning (ML)TermComputers learning patterns from data instead of following only fixed rules
- Deep learningTermMachine learning with many-layered neural networks
- Neural networkTermA system of connected nodes inspired loosely by brain cells
- Neuron (node)TermA unit in a network that takes inputs and gives an output
- LayerTermA group of nodes in a neural network
- WeightsTermNumbers in a network that are adjusted during learning
- ParametersTermThe values a model learns, such as weights
- TrainingTermTeaching a model using data
- Training dataTermExamples the model learns from
- Test dataTermNew examples used to check how well a model works
- Validation dataTermData used to tune a model during development
- LabelTermThe correct answer attached to an example
- FeatureTermA measurable property used as input
- ModelTermThe result of training; it makes predictions
- Prediction (inference)TermThe model's output on new data
- Supervised learningTermLearning from labeled examples
- Unsupervised learningTermFinding patterns in unlabeled data
- Reinforcement learningTermLearning by trial and error with rewards
- ClassificationTermSorting into categories, like cat or dog
- RegressionTermPredicting a number, like a house price
- ClusteringTermGrouping similar items together
- OverfittingTermA model that memorizes training data and does poorly on new data
- UnderfittingTermA model too simple to capture patterns
- Bias (in AI)TermUnfair results caused by skewed data or design
- FairnessTermMaking AI treat people equitably
- AlgorithmTermStep-by-step instructions
- Gradient descentTermA method for adjusting weights to reduce errors
- Loss functionTermA measure of how wrong the model is
- EpochTermOne full pass through the training data
- Learning rateTermHow big the steps are when updating weights
- AccuracyTermThe share of predictions that are correct
- Confusion matrixTermA table of correct and wrong predictions by class
- Computer visionTermAI that understands images and video
- Natural language processing (NLP)TermAI that works with human language
- Speech recognitionTermTurning speech into text
- Text-to-speechTermTurning text into spoken sound
- ChatbotTermA program that converses with people
- Large language model (LLM)TermA very large model trained on lots of text to predict and generate language
- TokenTermA piece of text, such as a word or part of a word, that language models process
- PromptTermThe input or instruction given to an AI model
- Prompt engineeringTermWriting prompts to get better answers
- Context windowTermHow much text a model can consider at once
- TransformerTermA neural network design behind most modern language models
- AttentionTermA mechanism that lets a model focus on relevant parts of its input
- Generative AITermAI that creates new text, images, audio or code
- Image generationTermCreating pictures from text descriptions
- Diffusion modelTermA model that creates images by gradually removing noise
- HallucinationTermWhen an AI states something false as if it were true
- Fine-tuningTermFurther training a model on specialized data
- Pre-trainingTermInitial training on a large general dataset
- EmbeddingTermA list of numbers representing the meaning of text or images
- Retrieval-augmented generation (RAG)TermLetting a model look up documents before answering
- AI agentTermAn AI system that can take actions toward a goal
- Turing testTermAlan Turing's idea of judging a machine by whether it can seem human in conversation
- Alan TuringTermBritish mathematician who asked 'Can machines think?' in 1950
- Dartmouth workshopTermThe 1956 meeting where the term artificial intelligence was coined
- Deep BlueTermIBM computer that beat chess champion Garry Kasparov in 1997
- AlphaGoTermDeepMind program that beat Go champion Lee Sedol in 2016
- ImageNetTermA huge image dataset that sparked deep learning progress
- Narrow AITermAI built for specific tasks
- General AI (AGI)TermHypothetical AI that can do any intellectual task a human can
- RoboticsTermBuilding machines that sense and act
- Self-driving carTermA vehicle that drives using sensors and AI
- Recommendation systemTermAI that suggests videos, songs or products
- DeepfakeTermA fake image or video made by AI
- AI ethicsTermThinking about fairness, safety, privacy and responsibility in AI
- Privacy and AITermAI systems should protect people's personal information
- Human in the loopTermA person reviews or guides the AI's decisions
- ExplainabilityTermUnderstanding why an AI made a decision
- Check AI answersTermAI can be wrong, so verify important facts
Download this deck
- CSVExcel, Google Sheets, Numbers
- Anki / Quizlet (tab-separated)Import straight into Anki or Quizlet
- JSONFor programmers and apps
- Plain textOne card a line, “front : back”, ready for Peabody's flashcard maker to print
Want paper cards? Paste the plain-text download into the flashcard maker.