History of AI Flashcards
Who is credited with designing the first programmable mechanical computer?
Charles Babbage, who designed the Analytical Engine in the 1840s.
What was the name of the first electronic general-purpose computer?
ENIAC (Electronic Numerical Integrator and Computer), built in the 1940s.
Who proposed the concept of the Universal Turing Machine?
Alan Turing, in 1936.
What was the name of the conference where the term “Artificial Intelligence” was coined?
The Dartmouth Conference, held in 1956
What are the two main approaches to AI?
Symbolic AI (Classical/GOFAI) and Connectionism/Neural Networks.
What is the main idea behind symbolic AI?
Symbolic AI relies on explicit symbolic representations like logical rules and semantic networks for knowledge representation and reasoning.
What is the main idea behind connectionist AI?
Connectionist AI, or neural networks, are inspired by the brain’s neural networks and use simplified artificial neuron models that activate based on input weights.
Who introduced the simplified mathematical model of an artificial neuron?
McCulloch and Pitts, in 1943
What was the name of the early neural network model developed by Frank Rosenblatt?
The Perceptron, introduced in 1957
What was the main limitation of early neural networks?
Early neural networks faced challenges in training and scalability.
What are the “AI Winters” referring to in the history of AI?
The “AI Winters” refer to periods of reduced funding and interest in AI research due to disillusionment with the lack of progress.
What are some factors that contributed to the revival of AI research?
Integration of ideas from cognitive science, use of probability theory, advances in neural network training algorithms, and increased availability of data and computing power.
What is the main difference between knowledge representation in symbolic AI and neural networks?
Symbolic AI uses explicit symbolic representations, while neural networks represent knowledge in a sub-symbolic, distributed way through connection weights.
What is the “grounding problem” in AI?
The grounding problem refers to the challenge of mapping or grounding abstract symbols in AI systems to the real world.
What are some advantages of neural networks compared to symbolic AI?
Neural networks can learn from data, handle uncertainty, and generalize better to novel patterns.
