LiZero transfers Q-value estimates from similar past MDPs into a new UCT-style bound (aUCT), claiming 3 to 4 times faster convergence and about 31% higher early rewards on a sequence of grid-world tasks.
Lipschitz Lifelong Reinforcement Learning
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
We consider the problem of knowledge transfer when an agent is facing a series of Reinforcement Learning (RL) tasks. We introduce a novel metric between Markov Decision Processes (MDPs) and establish that close MDPs have close optimal value functions. Formally, the optimal value functions are Lipschitz continuous with respect to the tasks space. These theoretical results lead us to a value-transfer method for Lifelong RL, which we use to build a PAC-MDP algorithm with improved convergence rate. Further, we show the method to experience no negative transfer with high probability. We illustrate the benefits of the method in Lifelong RL experiments.
fields
cs.AI 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Lipschitz Lifelong Monte Carlo Tree Search for Mastering Non-Stationary Tasks
LiZero transfers Q-value estimates from similar past MDPs into a new UCT-style bound (aUCT), claiming 3 to 4 times faster convergence and about 31% higher early rewards on a sequence of grid-world tasks.