オンライン学習 Protected: Trade-off between exploration and utilization -Regret and stochastic optimal measures, heuristics
Reinforcement learning with regrets, stochastic optimal measures, and heuristics
Reinforcement Learning
オンライン学習 オンライン学習 Protected: Planning Problems (2) Implementation of Dynamic Programming (Value Iterative Method and Measure Iterative Method)
Implementation of Dynamic Programming (Value Iteration and Policy Iteration) for Planning Problems as Reinforcement Learning for Digital Transformation , Artificial Intelligence and Machine Learning Tasks
強化学習 Protected: Planning Problems(1) – Approaches Using Dynamic Programming and Theoretical Underpinnings
Reinforcement learning by planning problems (dynamic programming and linear programming) for sequential decision problems in known environments used for digital transformation , artificial intelligence and machine learning tasks.
機械学習:Machine Learning Machine Learning Professional Series “Reinforcement Learning” Reading Memo
A reference book on reinforcement learning to observe the current situation and determine what action to take, used in digital transformation ,artificial intelligence , and machine learning tasks.
オンライン学習 Online learning and online prediction
Online learning is a sequential machine learning technique used in digital transformation , artificial intelligence , and machine learning tasks, and online prediction combines these techniques with decision-making problems.
微分積分:Calculus This is a good introduction to deep learning (Machine Learning Startup Series)Reading Notes
Overview of deep learning for digital transformation and artificial intelligence tasks, including machine learning, gradient descent, regularization, error back propagation, self-encoders, convolutional neural networks, recurrent neural networks, Boltzmann machines, and reinforcement learning.
数学:Mathematics Decision Theory and Mathematical Decision Making Techniques
We will discuss mathematical decision-making techniques used in reinforcement learning, online prediction, and algorithms for high-speed automated stock trading. After describing the four decision strategies, I will introduce subjective probability, Bayesian theory, multiple concept theory, and supply rise theory.
アルゴリズム:Algorithms Protected: Reinforcement Learning Overview
An overview of reinforcement learning for learning sequential decision rules