2012-10-10 hoelzl 2012-10-10 rule to show that conditional mutual information is non-negative in the continuous case
2012-10-10 hoelzl 2012-10-10 continuous version of entropy_le
2012-10-10 hoelzl 2012-10-10 simplified entropy_uniform
2012-10-10 hoelzl 2012-10-10 tuned product measurability
2012-04-23 hoelzl 2012-04-23 reworked Probability theory
2012-03-13 wenzelm 2012-03-13 prefer abs_def over def_raw;
2012-02-28 wenzelm 2012-02-28 avoid undeclared variables in let bindings; tuned proofs;
2011-12-07 hoelzl 2011-12-07 remove unnecessary sublocale instantiations in HOL-Probability (for clarity and speedup); remove Infinite_Product_Measure.product_prob_space which was a duplicate of Probability_Measure.product_prob_space
2011-12-01 hoelzl 2011-12-01 moved theorems about distribution to the definition; removed oopsed-lemma
2011-12-01 hoelzl 2011-12-01 remove duplicate theorem setsum_real_distribution
2011-09-12 nipkow 2011-09-12 new fastforce replacing fastsimp - less confusing name
2011-07-19 hoelzl 2011-07-19 Rename extreal => ereal
2011-06-27 hoelzl 2011-06-27 move conditional expectation to its own theory file
2011-06-09 hoelzl 2011-06-09 lemma: independence is equal to mutual information = 0
2011-03-29 hoelzl 2011-03-29 rename Probability_Space to Probability_Measure
2011-03-22 hoelzl 2011-03-22 standardized headers
2011-03-14 hoelzl 2011-03-14 reworked Probability theory: measures are not type restricted to positive extended reals
2011-02-23 hoelzl 2011-02-23 add lemma KL_divergence_vimage, mutual_information_generic
2011-02-02 hoelzl 2011-02-02 the measure valuation is again part of the measure_space type, instead of an explicit parameter to the locale; changed syntax for simple_function, simple_integral, positive_integral, integral and RN_deriv. introduced binder variants for simple_integral, positive_integral and integral.
2011-01-24 hoelzl 2011-01-24 use pre-image measure, instead of image
2010-12-29 wenzelm 2010-12-29 explicit file specifications -- avoid secondary load path;
2010-12-08 hoelzl 2010-12-08 cleanup bijectivity btw. product spaces and pairs
2010-12-03 hoelzl 2010-12-03 it is known as the extended reals, not the infinite reals
2010-12-01 hoelzl 2010-12-01 Support product spaces on sigma finite measures. Introduce the almost everywhere quantifier. Introduce 'morphism' theorems for most constants. Prove uniqueness of measures on cut stable generators. Prove uniqueness of the Radon-Nikodym derivative. Removed misleading suffix from borel_space and lebesgue_space. Use product spaces to introduce Fubini on the Lebesgue integral. Combine Euclidean_Lebesgue and Lebesgue_Measure. Generalize theorems about mutual information and entropy to arbitrary probability spaces. Remove simproc mult_log as it does not work with interpretations. Introduce completion of measure spaces. Change type of sigma. Introduce dynkin systems.
2010-09-13 nipkow 2010-09-13 renamed lemmas: ext_iff -> fun_eq_iff, set_ext_iff -> set_eq_iff, set_ext -> set_eqI
2010-09-07 nipkow 2010-09-07 expand_fun_eq -> ext_iff expand_set_eq -> set_ext_iff Naming in line now with multisets
2010-09-02 hoelzl 2010-09-02 Moved lemmas to appropriate locations
2010-09-02 hoelzl 2010-09-02 move lemmas to correct theory files
2010-08-23 hoelzl 2010-08-23 Rewrite the Probability theory. Introduced pinfreal as real numbers with infinity. Use pinfreal as value for measures. Introduces Lebesgue Measure based on the integral in Multivariate Analysis. Proved Radon Nikodym for arbitrary sigma finite measure spaces.
2010-05-04 hoelzl 2010-05-04 Corrected imports; better approximation of dependencies.
2010-05-03 hoelzl 2010-05-03 Cleanup information theory
2010-05-03 hoelzl 2010-05-03 Moved Convex theory to library.
2010-04-07 hoelzl 2010-04-07 Added Information theory and Example: dining cryptographers