Barto, A., and Mahadevan, S. 2003. Recent advances in hierarchical reinforcement learning. Discrete Event Dynamic Systems: Theory and Applications 13:41--77. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
COllective INtelligence with Sequences of Actions: Coordinating .. - Hoen, Bohte (2003) (3 citations) (Correct)
....heuristics and tuned system parameters such that hopefully the desired type of behavior emerges from running the system. Only recently has there been work on more top down type of approaches to establish the conditions for MASs such that they are most likely to exhibit good emergent behavior [1, 4, 2]. In typical problem settings, individual agent in the MAS contribute to some part of the collective through its private actions. The joint actions of all agents derive some reward from the outside world. To enable local learning, this reward has to be divided amongst the individual agents where ....
....of the agent in the grid and the action space consists the four directions in which the agent can move. The policy # of an agent in [10] is stochastic according to a softmax function; in the policy, a random action a i is chosen for state s and constant c (set at 50) with normalized chance in [0, 1] of c Q(s,a i ) j c Q(s,a j ) The discount factor # is set to 0.95. The learning rate # t at time step t for a state s is taken as # t = 1 0.0002#visits(s) where visits(s) is the number of times the state s has been visited during a learning step ( 6] The decreasing value of # serves ....
A. Barto and S. Mahadevan. Recent advances in hierarchical reinforcement learning. DiscreteEvent Systems journal, 2003. to appear.
COllective INtelligence with Sequences of Actions: Coordinating .. - Hoen, Bohte (2003) (3 citations) (Correct)
....heuristics and tuned system parameters such that hopefully the desired type of behavior emerges from running the system. Only recently has there been work on more top down type of approaches to establish the conditions for MASs such that they are most likely to exhibit good emergent behavior [1,4,2]. In typical problem settings, individual agent in the MAS contribute to some part of the collective through its private actions. The joint actions of all agents derive some reward from the outside world. To enable local learning, this reward has to be divided amongst the individual agents where ....
....of the agent in the grid and the action space consists the four directions in which the agent can move. The policy # of an agent in [10] is stochastic according to a softmax function; in the policy, a random action a i is chosen for state s and constant c (set at 50) with normalized chance in [0, 1] of c Q(s,a i ) j c Q(s,a j ) The discount factor # is set to 0.95. The learning rate # t at time step t for a state s is taken as # t = 1 0.0002#visits(s) where visits(s) is the number of times the state s has been visited during a learning step ( 6] The decreasing value of # serves ....
A. Barto and S. Mahadevan. Recent advances in hierarchical reinforcement learning. Discrete-Event Systems journal, 2003. to appear.
Internal Models and Anticipations in Adaptive Learning.. - Butz, Sigaud, Gérard (2 citations) (Correct)
....inputs that allow the identi cation of states. This ungeneralized exhaustive state representation prevents RL to scale up to larger problems. Several approaches exist that try to overcome the curse of dimensionality by function approximation techniques (cf. 52] hierarchical approaches (cf. [54, 4]) or online generalization mechanisms. Approaches that generalize online over sensory inputs (for example in the form of a feature vector) are introduced in the following. 4.2 Learning Classi er Systems Learning Classi er Systems (LCSs) have often been overlooked in the research area of RL ....
Barto, A.G., Mahadevan, S.: Recent advances in hierarchical reinforcement learning. Discrete event systems (2003, to appear)
Intrinsically Motivated Reinforcement Learning: A.. - Stout, Konidaris, Barto (2005) Self-citation (Barto) (Correct)
No context found.
Barto, A. G., and Mahadevan, S. 2003. Recent advances in hierarchical reinforcement learning. Discrete Event Dynamic Systems.
Learning and Approximate Dynamic Programming - Scaling.. - Si, Barto, Powell.. Self-citation (Barto) (Correct)
No context found.
A. Barto, and S. Mahadevan. Recent Advances in Hierarchical Reinforcement Learning. Discrete-Event Systems: Theory and Applications, 13:41--77, 2003.
Learning to Exploit Dynamics for Robot Motor Coordination - Rosenstein (2003) Self-citation (Barto) (Correct)
No context found.
A. G. Barto and S. Mahadevan. Recent advances in hierarchical reinforcement learning. Discrete Event Dynamic Systems: Theory and Applications, in press.
A Computational Model of the Cerebral Cortex - Dean (2005) (Correct)
No context found.
Barto, A., and Mahadevan, S. 2003. Recent advances in hierarchical reinforcement learning. Discrete Event Dynamic Systems: Theory and Applications 13:41--77.
Spoken Dialogue Management Using Hierarchical Reinforcement.. - Cuayįhuitl (2005) (Correct)
No context found.
Barto, A., and Mahadevan, S. (2003). Recent Advances in Hierarchical Reinforcement Learning. In Discrete Event Dynamic Systems: Theory and Applications, Kluwer Academic Publishers, 13, pp. 343-379.
Hierarchical Reinforcement Learning in.. - Fischer, Rovatsos, Weiss (2004) (Correct)
No context found.
A. G. Barto and S. Mahadevan. Recent advances in hierarchical reinforcement learning. Discrete Event Dynamic Systems, 13(4):41--77, 2003.
Analyzing Multi-Agent Reinforcement Learning Using Evolutionary .. - Hoen, Tuyls (Correct)
No context found.
A. Barto and S. Mahadevan. Recent advances in hierarchical reinforcement learning. Discrete-Event Systems journal, Special issue on Reinforcement Learning, 13:41--77, 2003.
Hierarchical Reinforcement Learning Based on Subgoal.. - Bakker, Schmidhuber (2004) (Correct)
No context found.
A. G. Barto and S. Mahadevan. Recent advances in hierarchical reinforcement learning. Discrete Event Systems, Special issue on reinforcement learning, 13:41--77, 2003.
Learning to Identify Irrelevant State Variables - Nicholas Jong University (2004) (Correct)
No context found.
Andrew G. Barto and Sridhar Mahadevan. Recent advances in hierarchical reinforcement learning. Discrete-Event Systems, 13:41--77, 2003. Special Issue on Reinforcement Learning.
Hierarchical Reinforcement Learning in.. - Fischer, Rovatsos, Weiss (Correct)
No context found.
A. G. Barto and S. Mahadevan. Recent advances in hierarchical reinforcement learning. Discrete Event Dynamic Systems, 13(4):41--77, 2003.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC