CIPP/E - AI question n answers graded A+ 2023/2024

CIPP/E - AIartificial intelligence - correct answers The capability of computer processing systems to perform functions otherwise defined as those requiring human intelligence to accomplish There are two categories of AI - correct answers general (or strong or full) AI and narrow (or weak or specialized) AI. General AI - correct answers The intelligence of a machine that could successfully perform any intellectual task that a human being can. The singularity - correct answers The moment at which machines become "smart" enough to take over their own development, and driving exponential or "runaway" advancement Narrow artificial intelligence - correct answers An application of artificial intelligence which replicates (or surpasses) human intelligence for a dedicated purpose or application. Machine Learning - correct answers A field of computer science which designs algorithmic based systems to learn from presented data rather than being explicitly programmed. A program or system that trains a predictive model from input data. Bias - correct answers The tendency of a process to over- or under-estimate the value (weight, influence) of one or more factors or parameters; may apply to the initial data collection or the processing applied to it Supervised Learning - correct answers Training a model from input data and its corresponding, previously assigned labels. Requires masses amounts of data and any bias present becomes inherent in the model it produces. Unsupervised learning - correct answers Training a model to find patterns using an unlabeled data set. reinforcement learning - correct answers A type of dynamic programming that trains algorithms using a system of reward and punishment. Uses no existing data for training. It may take longer to reach a "trained" state than other methods, but eliminates teacher bias Neural networks - correct answers A variation of a machine learning model that, taking inspiration from the human brain, is composed of multiple processing layers. Deep learning - correct answers A subfield of machine learning that uses a collection or sequence of algorithms to model higher-level abstractions in data. Based on neural network models with more than one hidden layer. Algorithm - correct answers A process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer. Explainability - correct answers The ability to describe or achieve human understanding of why a particular ML model arrived at a specific output or prediction. Transparency - correct answers The ability to explain or understand the inputs, outputs, functions, and data collection and use of a particular process, including the rights and relationship to the data subject and user.