Scaling Up Reinforcement Learning for Robot Control

By Longxin Lin. 1993

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Off-Policy Deep Reinforcement Learning by Bootstrapping the Covariate Shift (2019, methodology): “While many algorithms learn in a purely online fashion, sample-efficient methods typically make use of past data, viewed either as a fixed dataset, or stored in a replay memory (Lin 1993, Mnih et al. 2015).”
Safe and Efficient Off-Policy Reinforcement Learning (2016, methodology): “Retrace( λ ) can learn from full returns retrieved from past policy data, as in the context of experience replay (Lin, 1993), which has returned to favour with advances in deep reinforcement learning (Mnih et al., 2015; Schaul et al., 2016).”
Model-Based Reinforcement Learning with an Approximate, Learned Model (1996, methodology): “Loop: a := a 0 ; s := s 0 ; if s 0 is the terminal state, go to 2, else go to 3 This algorithm is similar to Experience Replay algorithm (Lin 1993) only instead of replaying a past experience we generate a new one using a model of the environment and the same action selection mechanism that we use…”
Q-Decomposition for Reinforcement Learning Agents (2003, background): “To overcome such problems, some have proposed command fusion , whereby the arbitrator executes some kind of combination (such as an average) of the subagents’ recommendations (Sa ﬃ otti et al., 1995; Ogasawara, 1993; Lin, 1993; Goldberg et al., in press).”
A Reinforcement Learning Technique with an Adaptive Action Generator for a Multi-robot System (2008, methodology): “Lin represented the Q-function by using multi-layer neural networks called Q-net (Lin, 1993).”
Symbiotic Evolution of Neural Networks in Sequential Decision Tasks (1997, methodology): “Q-learning has been shown to perform comparably to the AHC approach in terms of training time in toy domains (Lin 1992) and preliminary research has been done to scale up Q-learning to practical tasks (Lin 1993; Littman and Boyan 1993).”