Intro to AI

What is supervised learning?

A type of machine learning where a model is trained on labeled data to map inputs to outputs.

What is the goal of supervised learning?

To train the model to make predictions and decisions based on provided data.

What is classification in machine learning?

The process of sorting data into predefined classes and applying learned characteristics to predict discrete labels.

What are examples of classification tasks?

Spam detection, disease detection, image classification, bank loan prediction.

What is regression in machine learning?

A technique used to predict a continuous numerical value based on given input.

What are examples of regression tasks?

Predicting house prices or stock prices.

What is unsupervised learning?

A type of machine learning where models are trained on unlabeled data to find patterns and structures.

What is clustering?

Grouping data points into clusters of similar data points.

What are examples of clustering applications?

Search engines, face recognition, targeted marketing, recommender systems.

What is association rule mining?

A technique used to discover interesting relationships or patterns between variables in large datasets.

What is reinforcement learning?

A type of learning where an agent is rewarded or penalized based on its actions to find the best outcome.

What are the three steps for machine learning?

Train the model with data, validate model performance, and test the model on unseen data.

What are neural networks?

Computational models aimed at mimicking the human brain's structure and functioning.

What are the components of a neural network?

Input layer, hidden layers, output layer.

How do you evaluate model performance?

By splitting data into training, validating, and test sets, and assessing error metrics.

What is accuracy rate?

The ratio of correctly predicted instances to the total instances.

What is recall/sensitivity?

The rate of correctly predicted positive instances over actual total positive instances.

What is precision?

The ratio of correctly predicted positive instances to total predicted positive instances.

What is specificity?

The correctly predicted negative rate over total negative instances.

What are some applications of generative AI?

Image/video generation, music generation, text generation, personalized treatment plans.

What is an ROC curve?

A graphical representation that plots the false positive rate against the true positive rate.

What is a chatbot?

A computer program designed to simulate conversations with human users.

What is natural language understanding?

The ability of a computer to understand text or speech input.

What are the key components of NLU?

Intent recognition and entity recognition.

What is the main goal of natural language processing (NLP)?

Enable computers to understand, interpret, and generate human language meaningfully.

What are the phases of NLP?

Lexical analysis, syntactic analysis, semantic analysis, discourse analysis, pragmatic analysis.

What is generative AI?

AI that generates new content like images, sounds, and text.

What are the types of generative models?

Explicit density models and implicit density models.

What are GANs?

Generative Adversarial Networks, a two-part model consisting of a generator and a discriminator.

What do autoencoders do?

Map input data to a lower-dimensional representation and reconstruct the original data.

What are deepfakes?

Synthetic media where a person’s likeness is replaced with another using AI models.

What are some ethical issues concerning deepfakes?

Privacy, fraud, damage to reputation, misinformation, manipulation, copyright.

What is deep learning?

A subset of machine learning that uses neural networks to mimic the human brain.

What is computer vision?

The ability of computers to derive information from images and visual inputs.

What are potential future developments in AI?

Customized chatbots, analyzing mass data in science, AI in foreign policy, and optimizing energy consumption.