AI and tech Glossary
From the basics of AI to the techniques we actually use, one entry each.
27 entries
- AI agentAI 에이전트A program that takes a goal, breaks it into steps and uses tools to carry it out.
- Artificial intelligence인공지능An umbrella term for machines doing work that used to require human judgement, perception or creation.
- Computer vision컴퓨터 비전The field concerned with what is in an image or video, and where.
- Deep learning딥러닝A branch of machine learning that stacks many layers of artificial neurons.
- Embedding임베딩A representation of text or images as a vector of numbers, where similar meanings sit closer together.
- Fine tuning파인튜닝Taking an already trained model and training it a little further on your own data.
- GPUGPUA processor built to run many identical calculations at once, used for deep learning training and inference.
- Hallucination환각When a model produces something untrue but plausible sounding, as if it were fact.
- Image segmentation이미지 세그멘테이션Separating the pixels that belong to each object in an image, answering where with an outline rather than a box.
- Inference추론Running a trained model to get a result, as opposed to training it.
- Large language model대규모 언어 모델A model trained on very large amounts of text to predict what comes next, used for summarizing, translating, writing and conversation.
- Machine learning머신러닝Instead of writing the rules, you let a system find them in data.
- Neural network인공 신경망A structure of layered units that multiply inputs by weights, add them up and pass the result on.
- Overfitting과적합When a model fits its training data well but fails on new data.
- Parameter파라미터The numbers a model learns. Their count is what people call model size.
- Prompt프롬프트The instruction given to a model: what to do, in what form, under what conditions.
- RAGRAGFetching relevant documents first and answering from them, rather than from memory alone.
- Reinforcement learning강화학습Learning by acting, receiving a reward, and adjusting towards actions that earn more.
- Speech recognition음성 인식Turning spoken language into text.
- Supervised learning지도학습Learning from data where each input comes with the right answer.
- Text to speech음성 합성Turning text into speech, including systems that imitate a particular voice from a short recording.
- Token토큰The smallest chunk of text a model works with, sometimes shorter and sometimes longer than a word.
- Training data학습 데이터The data a model learns from. Its quality and bias decide the outcome.
- Transformer트랜스포머An architecture built on attention, which lets each word decide which other words to look at. It underlies most language and image models today.
- Unsupervised learning비지도학습Finding structure or groupings in data without any labels.
- Vision language model비전 언어 모델A model that handles images and text together, so a picture and a word can be compared for fit.
- Zero shot제로샷Asking a model to do a task it was not specifically trained for, using only a description.