Reinforcement Learning Using Neural Networks, with Applications to Motor Control

By Rémi Coulom. 2002

115 citations recorded by the source corpus.

How researchers see this paper

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

In Reinforcement Learning

Citations in the corpus

Benchmarking Deep Reinforcement Learning for Continuous Control (2016, methodology): “We implement ﬁve basic tasks that have been widely analyzed in reinforcement learning and control literature: Cart-Pole Balancing (Stephenson, 1908; Donaldson, 1960; Widrow, 1964; Michie & Chambers, 1968), Cart-Pole Swing Up (Kimura & Kobayashi, 1999; Doya, 2000), Mountain Car (Moore, 1990),…”
Learning Continuous Control Policies by Stochastic Value Gradients (2015, background): “Swimmer is a chain body with multiple links immersed in a fluid environment with drag forces that allow the body to propel itself [5, 27].”
DeepMind Control Suite (2018, methodology): “This procedurally generated k -link planar swimmer is based on (Coulom, 2002) but using MuJoCo’s high-Reynolds ﬂuid drag model.”
Gaussian Processes for Data-Efficient Learning in Robotics and Control (2015, neutral): “, without human demonstrations or known dynamics models [11], [18], [27], [45], [56].”
Reinforcement Learning by Value Gradients (2008, methodology): “This is described by (White and Sofge, 1992, ch. 10) and (Coulom, 2002, Appendix A).”
Autonomous blimp control using model-free reinforcement learning in a continuous state and action space (2007, methodology): “Therefore, function approximation techniques that directly operate on the continuous space such as neural networks [1], [15], kernel methods [9], or Gaussian processes [13], [3] have been proposed as powerful alternatives to the discrete approximations of the continuous Q-function.”
Thinking While Moving: Deep Reinforcement Learning with Concurrent Control (2020, background): “Finite differences can approximate the Bellman update in continuous-time stochastic control problems (Munos & Bourgine, 1998) and continuous-time temporal difference learning methods (Doya, 2000) can utilize neural networks as function approximators (Coulom, 2002).”
Harms from Increasingly Agentic Algorithmic Systems (2023, background): “Prior to 2013, RL systems were developed mainly for a restricted set of simple domains [Coulom 2002].”
PILCO: A Model-Based and Data-Efficient Approach to Policy Search (2011, methodology): “5 shows the interaction time with the cart-pole system required by pilco and algorithms in the literature that solved this task from scratch (Kimura & Kobayashi, 1999), (Doya, 2000), (Coulom, 2002), (Wawrzynski & Pacut, 2004), (Riedmiller, 2005), (Raiko & Tornio, 2009), (van Hasselt, 2010).”
Receding Horizon Differential Dynamic Programming (2007, methodology): “In this section we describe our model of the d-link swimmer dynamical system, first presented by Coulom [3].”
Value-gradient learning (2012, background): “4 of Coulom (2002). These also have asymptotic timings of O(dim(~ w) dim(~x)).”
A Practical and Conceptual Framework for Learning in Control (2010, background): “Classical references for kriging are the books by Matheron (1973), Cressie (1993), and Stein (1999). O’Hagan (1978) first describes GPs as a non-parametric prior over functions. Neal (1996, Chapter 2.1) shows that a neural network converges to a GP if the number of hidden units tends to infinity and”
Neural Reinforcement Learning Controllers for a Real Robot Application (2007, unknown): “In contrast to other work that applies RL to real world systems, we do not make use of a model or simulator to train our controllers (like e.g. in [1]).”
dm_control: Software and Tasks for Continuous Control (2020, background): “The physical model conforms to (Coulom, 2002) rather than the earlier Spong 1995.”