3 citations · 4 across the 6 of their papers we have counts for
4 papers · 1 filter
Average-DICE: Stationary Distribution Correction by Regression
Fengdi Che, Bryan Chan, Chen Ma +1
Off-policy policy evaluation (OPE), an essential component of reinforcement learning, has long suffered from stationary state distribution mismatch, undermining both stability and…
Target Networks and Over-parameterization Stabilize Off-policy Bootstrapping with Function Approximation
Fengdi Che, Chenjun Xiao, Jincheng Mei +6
We prove that the combination of a target network and over-parameterized linear function approximation establishes a weaker convergence condition for bootstrapped value estimation…
Correcting discount-factor mismatch in on-policy policy gradient methods
Fengdi Che, Gautham Vasan, A. Rupam Mahmood
The policy gradient theorem gives a convenient form of the policy gradient in terms of three factors: an action value, a gradient of the action likelihood, and a state distribution…
Bayesian Q-learning With Imperfect Expert Demonstrations
Fengdi Che, Xiru Zhu, Doina Precup +2
Guided exploration with expert demonstrations improves data efficiency for reinforcement learning, but current algorithms often overuse expert information. We propose a novel algor…