Theory Formation by Heuristic Search

By D. Lenat. 1983

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep learning in neural networks: An overview (2014, background): “thods of syntactic pattern recognition (Fu, 1977) such as grammar induction discover hierarchies of formal rules to model observations. The partially (un)supervised Automated Mathematician / EURISKO (Lenat, 1983; Lenat and Brown, 1984) continually learns concepts by combining previously learnt conce”
Formal Theory of Creativity, Fun, and Intrinsic Motivation (1990–2010) (2010, background): “Lenat's discovery system EURISKO [41], [42] has a preprogrammed interestingness measure which was observed to become more an more inappropriate ( " stagnation " problem) as EURISKO created new concepts from old ones with the help of human intervention.”
Goedel Machines: Self-Referential Universal Problem Solvers Making Provably Optimal Self-Improvements (2003, background): “Similar drawbacks hold for Lenat’s human-assisted, non-autonomous, self-modifying learner [21], our Meta-Genetic Programming [29] extending Cramer’s Genetic Programming [8, 1], our metalearning economies [29] extending Holland’s machine learning economies [14], and gradient-based metalearners for co”
Maximizing Fun by Creating Data with Easily Reducible Subjective Complexity (2013, background): “Lenat's discovery system EURISKO (Lenat 1983) has a preprogrammed interestingness measure which was observed to become more and more inappropriate ( " stagnation " problem) as EURISKO creates new concepts from old ones with the help of human intervention.”
Gödel Machines: Fully Self-referential Optimal Universal Self-improvers (2007, background): “Similar drawbacks hold for Lenat’s human-assisted, nonautonomous, self-modifying learner [22], our Meta-Genetic Programming [32] extending Cramer’s Genetic Programming [8, 1], our metalearning economies [32] extending Holland’s machine learning economies [15], and gradient-based metalearners for con”
Artificial curiosity based on discovering novel algorithmic predictability through coevolution (1999, background): “The central questions are: In a given environment, how to extract the predictable concepts corresponding to algorithmic regularities that are not already known? Which novel input sequence-transforming algorithms do indeed compute reduced internal representations permitting reliable predictions? Usua”
Simple Principles of Metalearning (1996, background): “For alternative views of metalearning, see, e.g., Lenat (1983), Rosenbloom et al. (1993).”
The Basic AI Drives (2008, background): “Eurisko was an AI system developed in 1976 [ 12 ] that could learn from its own actions.”
Shifting Inductive Bias with Success-Story Algorithm, Adaptive Levin Search, and Incremental Self-Improvement (1997, background): “, Lenat (1983), Rosenbloom et al. (1993). For instance, Lenat’s approach requires occasional human interaction defining the “interestingness” of concepts to be explored.”