Probability theory: the logic of science

By E. Jaynes, T. Grandy, Ray Smith, Tom Loredo, M. Tribus, John Skilling, G. Larry Bretthorst. 2005

Subfields: Probabilistic & Bayesian Methods, Statistical Learning Theory

How researchers see this paper

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

In Probabilistic & Bayesian Methods

In Statistical Learning Theory

Citations in the corpus

Meta-Learning Probabilistic Inference for Prediction (2018, methodology): “Appendix A provides a generalized derivation of the framework, grounded in Bayesian decision theory (Jaynes, 2003).”
Taming the Noise in Reinforcement Learning via Soft Updates (2015, background): “This prior policy has maximal entropy given the prior knowledge [20].”
Entropy Search for Information-Efficient Global Optimization (2011, background): “A functional that is well-deﬁned on probability densities and preserves many of the information-content interpretations of entropy (Jaynes and Bretthorst, 2003) is relative entropy , also known as Kullback-Leibler divergence (Kullback and Leibler, 1951).”
Unsupervised object discovery and localization in the wild: Part-based matching with bottom-up region proposals (2015, background): “In particular, in our setting with region proposals, foreground part and object regions cast votes for each other and make the regions ob- tain high confidences all together.”
Meta-learning of Sequential Strategies (2019, background): “pointing out that the Bayesian solution is typically viewed as following directly from an interpretation of probabilities as degrees of belief rather than from an optimization problem [Jaynes, 2003].”
Sigma-point kalman filters for probabilistic inference in dynamic state-space models (2004, background): “Probability theory provides a language for representing these uncertain beliefs and a calculus for manipulating these beliefs in a consistent manner [33, 89, 151].”
ON CUMULATIVE RESIDUAL EXTROPY (2020, background): “[12] provided a completion to theories of information based on entropy, resolving a longstanding question in its axiomatization as proposed by Shannon [24] and followed by Jaynes [11].”
Using Bayes to get the most out of non-significant results (2014, methodology): “The Bayes factor makes use of the full range of predictions of the alternative hypothesis, and can use this information to most sensitively draw conclusions (conditional on the information it assumes; Jaynes, 2003).”
You May Believe You Are a Bayesian But You Are Probably Wrong (2011, background): “This is particularly associated with Laplace’s principle of insufficient reason but received a more modern impetus with the appearance of the influential book on probability by Harold Jeffreys (1891–1989) (Jeffreys 1961; the first edition was in 1939) and also more recently with the work of Edwin Ja”
Probabilistic model identification of the bit–rock-interaction-model uncertainties in nonlinear dynamics of a drill-string (2010, methodology): “Such a nonparametric probabilistic approach uses the random matrix theory for which the probability distributions are constructed using the Maximum Entropy Principle (Jaynes (2003)).”
Bayesian surprise attracts human attention (2005, unknown): “In probability and decision theory it can be shown that the only consistent and optimal way for modeling and reasoning about uncertainty is provided by the Bayesian theory of probability [26, 27, 28].”
PyMC: Bayesian Stochastic Modelling in Python (2010, background): “The Bayesian interpretation of probability is epistemic, meaning random variable x's probability distribution p(x) represents our knowledge and uncertainty about x's value (Jaynes 2003).”
Reputation and Status as Motives for War (2014, background): “Formally, we can 7Technically, “correspondence with common sense” requires the weak condition that information that increases the plausibility of A and does not change the plausibility of B|A weakly increases the plausibility of AB and decreases the plausibility of not A ( Jaynes 2003, p. 14).”
Adaptive Posterior Learning: few-shot learning with a surprise-based memory module (2019, methodology): “This sequential algorithm was inspired by Bayesian posterior updating (Jaynes, 2003) in the sense that the output probability distribution is updated as more data is observed.”
Microbiome Datasets Are Compositional: And This Is Not Optional (2017, methodology): “…since this accounts for the precision of the estimate of the probabilities for each feature; we refer advanced readers to the more technical literature (Jaynes and Bretthorst, 2003; Fernandes et al., 2013; Gloor et al., 2016a) and the “Probability” section of the Supplement for more information.”