Introduction to Applied Artificial Intelligence

These flashcards cover key concepts from the lecture on applied artificial intelligence, focusing on agentic AI in finance, AI salaries, learning vs programming distinctions, statistical learning, model accuracy, KNN, model interpretability, overfitting, and evaluation metrics for machine learning models.

What does Agentic AI represent in the financial services industry?

It represents a leap beyond task automation to systems that can reason, plan, and act autonomously in alignment with risk, compliance, and trust requirements.

What is the projected median salary for AI professionals in the U.S. in 2025?

The median salary is projected to be $160,000, with entry-level positions starting at $70,000 to $120,000.

What is the main difference between learning and programming in the context of AI?

Learning involves creating a model that generalizes from examples, while programming involves crafting specific rules to represent knowledge.

What are the two main types of machine learning approaches?

Supervised learning and unsupervised learning.

Define statistical learning.

Statistical learning involves methods for uncovering patterns and relationships in data to make predictions or discover structure.

What are the main goals of assessing model accuracy?

To quantify how well predictions match observed data and to estimate how well a model will perform on unseen data.

What is k-nearest neighbors (KNN) in AI?

KNN is an algorithm that classifies data points by looking at the nearest 'k' data points and assigning the label based on the majority vote.

What does the trade-off between prediction accuracy and model interpretability imply?

Complex models may provide higher accuracy but lower interpretability, while simpler models may have higher interpretability but sacrifice accuracy.

What is overfitting in machine learning?

Overfitting occurs when a model learns the details and noise in the training data to the extent that it performs poorly on new data.

How does one evaluate model performance?

Model performance can be evaluated using metrics such as Mean Squared Error (MSE) for regression and accuracy, precision, and recall for classification.