B. Hengst

Publications
Influence

Discovering Hierarchy in Reinforcement Learning with HEXQ

B. Hengst
Computer Science
ICML
8 July 2002

An open problem in reinforcement learning is discovering hierarchical structure. Expand

Omnidirectional Locomotion for Quadruped Robots

B. Hengst, Darren Ibbotson, Son Bao Pham, C. Sammut
Computer Science
RoboCup
2001

Competing at the RoboCup 2000 Sony legged robot league, the UNSW team won both the challenge competition and all soccer matches, emerging the outright winners for this league against eleven international teams. Expand

Discovering hierarchy in reinforcement learning

B. Hengst
Mathematics, Computer Science
2003

This thesis addresses the open problem of automatically discovering hierarchical structure in reinforcement learning. Expand

Robot Localisation Using Natural Landmarks

Peter Anderson, Yongki Yusmanthia, B. Hengst, A. Sowmya
Computer Science
RoboCup
18 June 2012

This paper introduces an optimised method for extracting natural landmarks to improve localisation during RoboCup soccer matches. Expand

Hierarchical Approaches

B. Hengst
Computer Science
Reinforcement Learning
2012

We use a four-room task as a running example to illustrate the various concepts and approaches, including algorithms that can automatically learn the hierarchical structure from interactions with the domain. Expand

Structural Abstraction Experiments in Reinforcement Learning

R. Fitch, B. Hengst, D. Suc, G. Calbert, J. Scholz
Computer Science
Australian Conference on Artificial Intelligence
5 December 2005

This paper presents classes of abstraction useful for scaling up RL and points towards promising future work in two directions – the manual use of structural abstractions, and their automatic discovery. Expand

Model Approximation for HEXQ Hierarchical Reinforcement Learning

B. Hengst
Computer Science
ECML
20 September 2004

We extend HEXQ with heuristics that automatically approximate the structure of the task hierarchy for hierarchical reinforcement learning. Expand

Robocup Standard Platform League

B. Hengst
2014

This report describes the development and implementation of the locomotion for the Nao H25 V4 robot as used by team rUNSWift from the University of New South Wales, Australia, for the 2014 RoboCup… Expand

Fast Monocular Visual Compass for a Computationally Limited Robot

Peter Anderson, B. Hengst
Computer Science
RoboCup
24 June 2013

This paper introduces an extremely computationally inexpensive method for estimating monocular, feature-based, heading-only visual odometry using a monocular camera and 1-dimensional (1D) SURF features. Expand

RoboCup-2000 The Fourth Robotic Soccer World Championships

T. Balch, M. Fujita, +6 authors T. Weigel

sisting of about 500 people, participated in RoboCup-2000, and about 5000 spectators watched the events. RoboCup-2000 showed dramatic improvement over past years in each of the existing robotic… Expand

Recommended Authors