Key Concepts in AI and Machine Learning

Hidden Markov Models (HMMs)

A statistical model where the system is a Markov process with hidden states.

Perceptron

An early type of artificial neuron model in machine learning.

John Searle

Philosopher who proposed the Chinese Room Argument to critique strong AI.

Oliver Selfridge's Pandemonium model

A model of perception using 'demons' to explain pattern recognition.

Chinese Room Argument

A thought experiment arguing that symbol manipulation alone doesn't equal understanding.

Systems Reply

A counter-argument suggesting the whole system (not just the person) could understand.

Brain Simulator Reply

An argument suggesting simulating the entire brain might produce understanding.

Robot Reply

Suggests understanding could arise from embodiment and sensory interaction.

Daniel Dennett

A philosopher known for cognitive science and 'intuition pumps.'

Intuition Pump

A thought experiment that helps visualize abstract concepts.

Computational Theory of Mind

The idea that cognition is a form of computation.

Ray Kurzweil

Futurist who predicts the technological singularity.

Singularity

The hypothetical point when AI surpasses human intelligence.

Moore's Law

Observation that computing power doubles approximately every two years.

Expert Systems

Rule-based programs designed to simulate expert human reasoning.

Edward Feigenbaum

Pioneer of expert systems in AI.

Speech Recognition

Technology that converts spoken language to text.

Geoffrey Hinton

Pioneer in neural networks and backpropagation.

Ilya Sutskever

Co-founder of OpenAI and a deep learning expert.

Generative AI

AI that creates new data (e.g., text, images, music).

Large Language Models (LLMs)

Machine learning models trained on massive text corpora.

Big Data

Extremely large data sets enabling modern AI techniques.

Deep Neural Networks

Multi-layered networks used for learning complex data patterns.

Supervised Learning

Model training with labeled input-output pairs.

Self-Supervised Learning

Training without external labels; models generate their own supervision.

Nick Bostrom

Philosopher focused on AI safety and future risks.

Paperclip Maximizer

A thought experiment about AI misalignment.

GPT-2

A language model developed by OpenAI.

OpenAI

AI research lab that created the GPT series and other models.

AI Winter

Periods when interest and funding in AI declined.

I.J. Good

Mathematician who coined 'intelligence explosion.'

Intelligence explosion

Concept introduced by I.J. Good in 1965.

Black boxed AI

Its decision process is not transparent or interpretable.

Vernor Vinge

Author who coined the term 'technological singularity.'

Unsupervised Learning

Modeling without labeled data to uncover hidden patterns.

DARPA's role in AI

Funds innovative AI research (e.g., NLP, robotics).

Brittleness in AI

The tendency of AI to fail outside narrow, trained domains.

Norbert Wiener

Founder of cybernetics.

Ada Lovelace

First computer programmer for the Analytical Engine.

Alan Turing

Father of theoretical computer science and AI.

Joseph Weizenbaum

Created the ELIZA program.

John McCarthy

AI pioneer who coined 'Artificial Intelligence.'

ELIZA

An early rule-based NLP program that mimicked a therapist.

Moravec's Paradox

Hard cognitive tasks are easy for AI, but motor tasks are hard.

Marvin Minsky

AI pioneer and co-founder of MIT AI Lab.

Dartmouth Workshop

The founding conference of AI.

Charles Babbage

Designed the Analytical Engine; called the father of the computer.

Turing Test

A test to see if a machine can exhibit human-like intelligence.

Turing Machine

Abstract model of computation described by Alan Turing.

Cybernetics

The study of control and communication in machines and organisms.

NLP (Natural Language Processing)

AI field concerned with understanding and generating human language.

Herb Simon

Psychologist and AI pioneer; co-developed GPS.

Alan Newell

Co-creator of early AI systems like the General Problem Solver.

General Problem Solver (GPS)

An early symbolic AI system developed in 1957.

Symbolic AI

AI based on manipulating symbols to represent logic and rules.

Automata

Abstract computational models like finite-state machines.

Perceptrons

1969 book by Minsky and Papert that criticized limitations of simple neural nets.

Differential Technological Development

Bostrom's strategy to guide tech growth in beneficial directions.

Frank Rosenblatt

Creator of the Perceptron model in 1958.