Главная
Study mode:
on
1
Intro
2
Model-Based Reinforcement Learning
3
Episodic Reinforcement Learning
4
Upper Confidence Model-Based RL (UCRL)
5
The class of deterministic continuous systems . Consider a deterministic system
6
A Simple Metric-Based RL Algorithm
7
Doubling Dimension d
8
Feature space embedding of transition model
9
The MatrixRL Algorithm
10
From Feature to Kernel Embedding of Transition Model
11
A motivating example: MuZero
12
Assumption of Value-Targeted Regression
13
Value-Targeted Regression (VTR) for Confidence Set Construction
14
Full Algorithm of UCRL-VTR
15
Regret analysis of UCRL-VTR
16
A Special Case
Description:
Explore model-based reinforcement learning with value-targeted regression in this 36-minute lecture by Mengdi Wang from Princeton University. Delve into episodic reinforcement learning, upper confidence model-based RL (UCRL), and deterministic continuous systems. Examine the MatrixRL algorithm, feature space embedding of transition models, and kernel embedding techniques. Investigate the motivating example of MuZero and the assumptions behind value-targeted regression. Learn about confidence set construction using VTR and analyze the regret of UCRL-VTR. Gain insights into the mathematics of online decision-making and advanced reinforcement learning concepts.

Model-Based Reinforcement Learning with Value-Targeted Regression

Simons Institute
Add to list
#Computer Science #Machine Learning #Reinforcement Learning #Model Based Reinforcement Learning
0:00 / 0:00
1 of 16

Intro