AWA AIF-C01 - Domain 1 — Fundamentals of AI and ML

The broad field of creating systems that perform tasks requiring human-like intelligence, such as reasoning, perception, language understanding, and decision-making.

A subset of AI where systems learn patterns from data to make predictions or decisions rather than being explicitly programmed for every outcome.

A type of machine learning that uses neural networks with multiple layers to learn complex patterns from large amounts of data.

A machine learning model inspired by biological neural networks, composed of interconnected nodes organized into layers.

AI that can create new content such as text, images, audio, video, or code based on patterns learned from data.

AI systems that can pursue goals by reasoning, planning, taking actions, and potentially using tools or external systems.

A trained computational representation that has learned patterns from data and can use those patterns to produce predictions, classifications, or other outputs.

A defined procedure or set of rules used to process data or solve a problem.

The process of using data to adjust a model so it learns patterns or relationships useful for a particular task.

Using a trained model to generate a prediction, classification, or other output from new input data.

Inference performed on a collection of data at once rather than processing each request immediately.

Inference where a prediction or response is generated immediately or with very low latency after a request.

Inference where the requester does not need to remain connected while processing occurs; the result is provided later.

Model inference using infrastructure managed by the cloud provider, allowing the user to focus on requests rather than managing servers.

Machine learning using labeled training data, where the desired output is known for each training example.

Machine learning that discovers patterns or structures in data without labeled target outputs.

Learning through interaction with an environment, using rewards or penalties to encourage desirable behavior.

Data that includes known target values or categories used to train supervised learning models.

Data without predefined target labels, commonly used for unsupervised learning and other exploratory techniques.

Data organized according to a defined schema, such as rows and columns in a relational database.

Data without a predefined tabular structure, such as documents, images, audio, and video.

Data organized into rows and columns, often containing structured features and values.

Data consisting of observations associated with points or intervals in time, often used for forecasting.

AI/ML techniques used to interpret and extract information from images and video.

AI techniques that enable computers to process, understand, analyze, and work with human language.

A machine learning task that assigns an input to one or more predefined categories.

A machine learning task that predicts a continuous numerical value, such as price, demand, or temperature.

An unsupervised learning technique that groups similar data points together without predefined categories.

An AI/ML system that predicts items, content, or actions that may be relevant to a user based on available data and patterns.

Using historical and other relevant data to predict future values or events, often involving time-series data.

Using AI/ML to identify transactions or behaviors that appear suspicious or inconsistent with expected patterns.

A systematic tendency in data, model behavior, or outcomes that can cause certain results to be favored or disadvantaged.

The degree to which an AI system's outcomes avoid unjustified or inappropriate differences across individuals or groups.

When a model learns the training data too closely, including noise, and performs poorly on new, unseen data.

When a model is too simple or insufficiently trained to capture important patterns in the data.

The sequence of stages involved in developing and operating an ML solution, from data preparation and training through deployment and monitoring.

Assessing how well a trained model performs using appropriate performance metrics and data that represents its intended use.

The proportion of predictions that are correct out of all predictions made.

The proportion of positive predictions that are actually positive. It focuses on minimizing false positives.

The proportion of actual positive cases that the model correctly identifies. It focuses on minimizing false negatives.