// LINE PRINTER OUTPUT — FORM FEED 0001
Foundational
Papers
in computer science & artificial intelligence
-
01
On Computable Numbers, with an Application to the Entscheidungsproblem
Alan Turing · 1936 · Proceedings of the London Mathematical Society
Defines the abstract machine — tape, states, a head that reads and writes — that every computer is still an instance of.
-
02
A Mathematical Theory of Communication
Claude Shannon · 1948 · Bell System Technical Journal
Turns information into a measurable quantity and names its unit: the bit.
-
03
The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain
Frank Rosenblatt · 1958 · Psychological Review
The first trainable artificial neuron — a machine that learns its weights from examples.
-
04
Perceptrons
Marvin Minsky & Seymour Papert · 1969 · MIT Press
Proves what a single-layer perceptron cannot learn — XOR among them — cooling the field for a decade. A book; no official free PDF.
-
05
Time, Clocks, and the Ordering of Events in a Distributed System
Leslie Lamport · 1978 · Communications of the ACM
Shows that “before” in a distributed system is about causal order, not wall-clock time.
-
06
Learning Representations by Back-Propagating Errors
Rumelhart, Hinton & Williams · 1986 · Nature
Backpropagation makes multi-layer networks trainable — the answer to the 1969 critique.
-
07
The Anatomy of a Large-Scale Hypertextual Web Search Engine
Sergey Brin & Lawrence Page · 1998 · Computer Networks
PageRank ranks a page by who links to it — the paper that became Google.
-
08
ImageNet Classification with Deep Convolutional Neural Networks
Krizhevsky, Sutskever & Hinton · 2012 · NeurIPS
A deep CNN trained on GPUs crushes ImageNet, igniting the modern deep-learning era.
-
09
Attention Is All You Need
Vaswani et al. · 2017 · Google Brain / NeurIPS
Replaces recurrence with attention — the Transformer behind today’s language models.
-
10
Language Models are Few-Shot Learners
Brown et al. · 2020 · OpenAI / NeurIPS
Scaling a Transformer to 175 billion parameters lets it learn new tasks from a few examples in the prompt.
END OF TAPE