2. Types of AI Flashcards
What are the three high level categories of artificial intelligence?
- artificial Narrow Intelligence (ANI)
- Artificial General Intelligence (AGI)
- Artificial Superintelligence (ASI)
What is Artificial Narrow Intelligence (ANI)?
- Designed to perform a single or a narrow set of related tasks at a high level of proficiency (ex. A system designed to play chess)
- Operates under a narrow set of constraints and limitations
- Commonly referred to as weak AI
- Boosts productivity and efficiency by automating repetitive tasks, enabling smarter decision-making and optimization through trend analysis
- Embedded in many industries, from health care to financial services, manufacturing and
customer service to benefit both organizations and end users
What is Broad Artificial Intelligence?
- More advanced in scope than Artificial Narrow Intelligence (ANI), capable of performing a broader set of tasks
- Relies on a group of AI systems, capable of working together and combining their
decision-making capabilities (ex. autonomous driving vehicles) - Lacks full, human-like capabilities experts expect of Artificial General Intelligence
What is Artificial General Intelligence (AGI), Strong, deep, or Full AI?
Systems that have strong generalization capabilities, the ability to think, understand, learn and preform complex tasks, and achieve goals in different contexts and environments. (Closely mimic human intelligence)
What is Artificial Superintelligence (ASI)?
AI systems with intellectual powers beyond those of humans across a comprehensive range of categories and fields of endeavor (Capable of outperforming humans, self-aware, understanding human emotions and capable of evoking its own - experiencing reality as a human)
What is machine learning?
- Branch of AI that leverages data and algorithms to enable systems to repeatedly learn and make decisions
- Improves over time without being explicitly instructed or programmed to do so.
What are the primary models of machine learning?
- Supervised learning
- Unsupervised learning
(Semi-supervised learning) - Reinforcement learning
Describe the Supervised Learning:
Model learns from pre-labeled and classified data set.
- An algorithm analyzes the input data and associated labels to produce an inferred
function, that becomes the basis for the system to make predictions based on new,
previously unseen inputs - Compare their outputs with correct or intended output, to identify errors and
improve prediction skills (ex. model that analyzes images of road signs labeled to define the sign’s meaning or purpose)
What are the two subcategories of the Supervised Learning Model?
- Classification models: produce outputs in the form of a specific categorical response (ex. whether an image contains a puppy)
- Regression models: predict a continuous value (ex. estimating a stock price)
Describe Unsupervised Learning:
Model does not rely on labeled datasets.
- Designed to identify differences, similarities and other patterns without human supervision
- More cost-efficient and require less effort but also susceptible to producing less accurate outputs and display unpredictable behaviors
What are two subcategories of Unsupervised Learning?
- Clustering: automatically grouping data points that share similar or identical attributes (ex. DNA patterns that share similarities or patterns)
- Association rule learning: identifying relationships and associations between data points (ex. understanding consumer buying habits, anomaly identification in mechanical faults, fraud identification, consumer segmentation and marketing strategies, genetics)
Describe Reinforcement Learning:
The use of reward and punishment matrix to determine a correct or optimal outcome
- Rely on trial and error to determine what to do or what not to do; rewarded or
punished accordingly - Do not ingest pre-labeled data sets; learning is solely through action and repetition,
changing or not changing state or getting feedback from their environment - Actions and decisions that result in a reward reinforce the triggering behavior
(incentivizing the model to follow the same tactic in the future) - Errors trigger a penalty and reduce the reward, proportional in size to the scale of
the error
Examples:
- A robot navigating a maze or stocking shelves in a large warehouse (robot is
rewarded for identifying the most efficient routes and avoiding certain obstacles or
hazards)
- Generative predictive text (making the model mimic human writing or
responses based on feedback)
- Improving the placement of online ads in a real-time bidding environment
- Amazon’s Warehouse Supply Chain Optimization
Describe semi-supervised Learning:
A combination of Supervised and Unsupervised Learning processes
- Uses a small amount of labeled data and a large amount of unlabeled data
- Aims to leverage the benefits of both models: improving reliability while reducing costs
- Helpful in scenarios where it is challenging to find or create a large pre-labeled dataset *
-Examples:
- Image and speech analysis
- Categorization and ranking of web page search results
- Large Language Models (LLMs): A form of AI that utilizes deep learning algorithms to
create models trained on massive text data sets to analyze and learn patterns and
relationships among characters, words and phrases. LLMs often rely on Semi-Supervised Learning Models
- ChatGPT, Dall-e and other generative AI tools
What is Machine Perception?
- Evolving field of potential convergence between robotics and AI
- Systems are trained to process sensory information and mimic the human senses: sight, sound, touch, smell and taste
- Robotic sensors provide relevant data through cameras, microphones, pressure sensors, 3D scanners, motion detectors and thermal imaging
- Combining sensors and AI models enable systems to sift through data at a much faster rate and order of magnitude beyond human ability; eliminating noise, analyzing and categorizing information
Examples: food production: improve preparation and storage by developing systems that can
touch, smell and taste the produce
What is Robotic Process Automation (RPA)?
- Evolving technology that uses software robots to automate repetitive and rule-based tasks within business processes
- Designed to mimic human actions on digital systems, (ex. data entry and form processing)
- Natural language processing machines learning enhances RPA robots’ automation capabilities
