Basic AI Concepts and Terminologies

Welcome, students!

In this lesson, we introduce fundamental artificial intelligence (AI) concepts and terminologies that explain why AI matters and how it is reshaping industries such as healthcare, finance, and retail. This foundational knowledge is also key for certification paths like the AWS Certified AI Practitioner.

The Impact of AI-Powered Applications

AI-powered applications are revolutionizing various fields by enhancing efficiency, reducing operational costs, and accelerating decision-making. Companies are adopting AI solutions—from chatbots to automated data analysis—to serve as force multipliers, enabling smarter and faster business operations.

The image highlights the importance of AI for organizations, emphasizing its applications in customer service, fraud detection, and data analysis.

The ultimate goal is to enhance customer experiences while decreasing the human workload. This dynamic shift has spurred substantial demand for AI skills across industries.

The image is an infographic titled "Why Artificial Intelligence (AI) Matters," featuring a robot icon and highlighting two benefits: enhancing customer experience and reducing human workload.

AWS and Its Role in Advancing AI

AWS is a major enabler of AI innovation with services such as SageMaker for machine learning and Rekognition for image analysis. These tools democratize access to advanced data processing and categorization, even for non-experts, breaking down previous barriers.

The image highlights the importance of artificial intelligence, featuring icons for Amazon SageMaker and Amazon Rekognition.

Transition to Machine Learning Systems

Since around 2015-2016, there has been a significant shift from traditional rule-based software to machine learning systems. Rule-based systems follow fixed logic—for instance, credit approval based on pre-set thresholds—whereas machine learning (ML) systems learn probabilistically from historical data, adapting to unseen scenarios.

The image compares rule-based systems with machine learning systems, featuring icons representing each type.

Note

While rule-based systems provide consistent outputs for straightforward tasks, machine learning offers superior capabilities for applications such as recommendations, predictions, and forecasts.

What Is Artificial Intelligence?

Artificial intelligence is a branch of computer science that creates systems capable of performing tasks that traditionally require human intelligence. Modern AI encompasses areas including visual perception, speech recognition, decision making, and language translation.

The image contains a definition of Artificial Intelligence (AI), describing it as a branch of computer science focused on creating systems capable of performing tasks that typically require human intelligence.

The image is an infographic titled "Artificial Intelligence (AI)" that highlights four key areas: visual perception, speech recognition, decision-making, and language translation.

Exploring AI Subfields: Machine Learning and Deep Learning

AI can be broadly divided into subfields such as Machine Learning (ML) and Deep Learning (DL):

The image is a diagram showing "Artificial Intelligence (AI)" with two components: "Machine Learning (ML)" and "Deep Learning (DL).

Narrow AI vs. General AI

Narrow AI: Focused on specific tasks such as product recommendations or personalized interactions (e.g., AI assistants like Alexa, Siri, or personalized content on platforms like Netflix).
General AI: Represents a theoretical system with broad problem-solving abilities akin to human intelligence. Although generative AI is making strides toward broader capabilities, true general AI remains a long-term goal.

The image illustrates "Narrow AI (specific tasks)" with examples like Alexa, Siri, and Netflix, represented by their logos.

General AI is distinct from generative AI, which primarily focuses on specific outputs. While generative AI is pushing boundaries towards flexible problem-solving, it is not yet synonymous with the expansive vision of general AI.

The image is a slide titled "General AI (broad capabilities)" with a description stating it is a theoretical form of AI with broad problem-solving abilities, similar to a human.

Machine Learning: Learning from Data

In practical applications, AI can perform binary classification tasks (e.g., identifying spam emails) or probabilistic predictions (e.g., forecasting market trends). This data-driven approach is known as machine learning.

ML employs mathematical algorithms and statistical models to identify patterns within both structured and unstructured data. Unlike traditional software that strictly executes programmed instructions, ML models continually improve as they process more data and receive feedback.

The image contains a definition of Machine Learning (ML), explaining it as the use of algorithms and statistical models to enable computers to learn from data rather than following explicit instructions.

For example, to train an ML model to differentiate between cats and dogs, you would use accurately labeled images. The model learns key features such as ear shape, snout structure, and eye characteristics, applying this knowledge to classify new images. Services like Gmail utilize these models to filter spam, and chatbots evolve based on interaction feedback.

The image is an infographic about Machine Learning (ML), highlighting the use of training data to build models and the identification of patterns for making predictions.

Deep Learning: A Specialized Subset

Deep learning, a subset of machine learning, leverages multi-layered neural networks. These networks, composed of interconnected artificial neurons, process data through multiple layers—similar to an intricately weighted decision tree—to solve complex problems.

The image explains that deep learning is a subset of machine learning (ML) using multi-layered neural networks to solve complex problems.

Consider an input such as a cat image: the neural network processes features like ear shape and eye structure across various layers, with each layer refining the confidence of the classification.

The image illustrates a neural network diagram with an input layer, multiple hidden layers, and an output layer, showing the connections between nodes.

Deep learning is well-suited for complex tasks such as real-time voice assistance, facial recognition, and language translation, owing to its ability to analyze intricate patterns with high accuracy.

Summary

Artificial intelligence is an overarching field that includes machine learning and deep learning. In essence:

AI refers to any technology that can replace or augment human effort.
Machine learning utilizes algorithms to learn from historical data.
Deep learning refines these capabilities using multi-layered neural networks that mimic human reasoning.

The image is a Venn diagram comparing Artificial Intelligence, Machine Learning, and Deep Learning, with brief descriptions of each concept.

These concepts form the foundation of modern AI. Embrace the journey into AI technologies to unlock new possibilities in enhancing decision-making and automating complex tasks.

Thank you for reading this article. We look forward to guiding you through the next lesson on advanced AI techniques.

Watch Video

Watch video content