In 1959, IBM researcher Arthur Samuel published “Some Studies in Machine Learning Using the Game of Checkers” in the IBM Journal of Research and Development. The paper described a checkers-playing program that improved its own play by learning from experience rather than being explicitly programmed with expert strategy, an early and influential demonstration of machine learning in action. Only the year of publication is documented, so this entry uses 1959-01-01.
Samuel’s program adjusted an evaluation function based on the outcomes of games, including games it played against itself, foreshadowing ideas central to modern reinforcement learning. The work is widely cited for popularizing the term “machine learning.”
The result was significant because it showed, at the dawn of the field, that a computer could get better at a nontrivial task on its own. It helped establish learning from data and experience as a serious approach to artificial intelligence.