グラフ理論

Protected: Computational Methods for Gaussian Processes(2)Variational Bayesian Method and Stochastic Gradient Method

Calculations using variational Bayesian and stochastic gradient methods for Gaussian process models, an application of stochastic generative models utilized in digital transformation, artificial intelligence, and machine learning tasks Kullback-Leibler information content, Jensen inequality, evidence lower bound function, mini-batch method, evidence lower bound, variational posterior distribution, evidence variational lower bound

2022.10.10

アルゴリズム:Algorithmsグラフ理論スパースモデリングベイズ推定微分積分:Calculus推論技術:inference Technology最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra

Protected: Maximum Propagation Method for Calculating MAP Assignments in Graphical Models

Estimating the maximized state of probability values (MAP assignment) with the maximum propagation method in probabilistic generative models used in digital transformation, artificial intelligence, and machine learningtasks (TRW maximum propagation method, STA condition, maximum propagation method on a factor graph with cycles, maximum propagation on a tree graph, MAP estimation by message propagation)

2022.09.28

アルゴリズム:Algorithmsグラフ理論ベイズ推定微分積分:Calculus推論技術:inference Technology数理論理学:Mathematical logic最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra

Protected: Foundations of Measure Theory for Nonparametric Bayesian Theory

Foundations of measure theory for nonparametric Bayesian theory (independence of random measures, monotone convergence theorem in Laplace functionals, propositions valid with probability 1, Laplace transform of probability distribution, expectation computation by probability distribution, probability distribution, monotone convergence theorem, approximation theorem by single functions, single functions, measurable functions using Borel set families, Borel sets, σ-finite measures, σ-additive families, Lebesgue measures, Lebesgue integrals)

2022.09.27

アルゴリズム:Algorithmsグラフ理論ベイズ推定微分積分:Calculus推論技術:inference Technology最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra

Protected: Calculation of Gaussian processes (1) Calculation by the auxiliary variable method

Approximate computation of Gaussian process models, an application of stochastic generative models utilized in digital transformation, artificial intelligence, and machine learning tasks, using the partial data method and the auxiliary variable method

2022.09.26

アルゴリズム:Algorithmsグラフ理論ベイズ推定微分積分:Calculus推論技術:inference Technology最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra

Modeling that combines probability and logic (2) PLL (Probabilistic Logical Learning)

Fusion modeling of probability and logic Probabilistic Logical Learning, ILP, PRISM used for digital transformation, artificial intelligence, and machine learning tasks.

2022.09.19

Symbolic Logicアルゴリズム:Algorithmsエキスパートシステム:expertsystemグラフ理論シミュレーションベイズ推定微分積分:Calculus推論技術:inference Technology数理論理学最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra集合論:Set theory

Protected: Replica Exchange Monte Carlo and Multicanonical Methods

On replica exchange Monte Carlo and multi-canonical methods, which are algorithms to avoid falling into local solutions in Markov chain Monte Carlo methods used for digital transformation, artificial intelligence, and machine learning tasks.

2022.09.18

アルゴリズム:Algorithmsグラフ理論ベイズ推定幾何学:Geometry微分積分:Calculus数理論理学:Mathematical logic最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra集合論:Set theory

Protected: Specific examples of graphical models

Computation of specific graphical models such as Boltzmann Machines, Mean Field Approximation, Bethe Approximation, Hidden Markov Models, Bayesian Hidden Markov Models, etc. as probabilistic generative models utilized in digital transformation, artificial intelligence and machine learning tasks.

2022.09.14

アルゴリズム:Algorithmsグラフ理論ベイズ推定微分積分:Calculus推論技術:inference Technology最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra

Protected: Nonparametric Bayesian Applications to Factor Analysis and Sparse Modeling

Nonparametric Bayesian models, one of the applications of probabilistic generative models utilized in digital transformation, artificial intelligence, and machine learning tasks, for factor analysis and sparse modeling (infinite latent feature model, beta-Bernoulli distribution model, Indian cuisine buffet process, binary matrix generation process)

2022.09.13

アルゴリズム:Algorithmsグラフ理論ベイズ推定微分積分:Calculus推論技術:inference Technology最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra集合論:Set theory

Protected: Stochastic Generative Models and Gaussian Processes(3) Representation of Probability Distributions

Stochastic generative models utilized in digital transformation, artificial intelligence, and machine learning tasks and representation of probability distributions in samples as a basis for Gaussian processes ,weighted sampling, kernel density estimation, distribution estimation using neural nets

2022.09.12

アルゴリズム:Algorithmsグラフ理論ベイズ推定微分積分:Calculus推論技術:inference Technology最適化:Optimization機械学習:Machine Learning確率・統計:Probability and Statistics線形代数:Linear Algebra

Protected: Support Vector Machines for Weak Label Learning (2) Multi-Instance Learning

Extension of support vector machines utilized for digital transformation, artificial intelligence, and machine learning tasks; multi-instance learning approach with SVMs for weak-label learning problems (mi-SVM, MI-SVM)

2022.09.02

アルゴリズム:Algorithmsグラフ理論微分積分:Calculus最適化:Optimization機械学習:Machine Learning画像認識技術異常検知・変化検知確率・統計:Probability and Statistics線形代数:Linear Algebra自然言語処理:Natural Language Processing音声信号認識技術