Bernhard Hengst Home Page

Bernhard Hengst

Research Group Manager

Making Sense of Data

Neville Roach Laboratory

bernhard.hengst@nicta.com.au

Current affiliations

NICTA

Compter Science and Engineering, University of New South Wales

Short Bio

Bernhard Hengst is Research Group Manager, Making Sense of Data research theme, at the Neville Roach Research Laboratory, NICTA at Kensington in Sydney. His research interests include machine intelligence and robotics. Prior to joining NICTA, Dr Hengst was General Manger at Ferntree Computer Corporation Limited in Australia and GE Capital IT Solutions in the USA. Previously he worked in Operations Research for ICI Limited in Australia and the UK. Dr Hengst holds a PhD from the University of New South Wales.

Research Interests

General purpose intelligent systems that can learn and be taught to perform many different tasks autonomously by interacting with their environment over a lifetime. Cognitive architectures that involve modular layered learning.

Publications & Reports

2008 (accepted)
NICTA & RTA Partnership for Traffic Control Innovation. Christian Chong-White, Bernhard Hengst, 23rd ARRB Conference, Adelaide, 2008.

2007
STaR Control - 2007 Report, Carlos Aydos, Robert Fitch, Glenn Geers, Bernhard Hengst, Eric Huang, George Katsirelos, Philip Kilby, Michael Maher, Paul Tyler, Toby Walsh, NICTA STaR Project, 2007.

Safe State Abstraction and Reusable Continuing Subtasks in Hierarchical Reinforcement Learning  Bernhard Hengst, Twentieth Australian Joint Conference on Artificial Intelligence - AI07, Gold Coast, Queensland, Australia December 2007.

Extracting Terrain Features from Range Images for Autonomous Random Stepfield Traversal, Raymond Sheh, M. Waleed Kadous, Claude Sammut, Bernhard Hengst, The IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR2007), September 2007.

Sensing, Perception and Action in Complex Environments (SPACE) Strategic Plan. Ed. Bernhard Hengst NICTA 2007

2006
STaR Control - 2006 Report, Robert Fitch, Bernhard Hengst, Philip Kilby, Michael Maher, Nobuyuki  Morioka, Paul Tyler, Toby Walsh. NICTA STaR Project, 2006.

2005

Structural Abstraction Experiments in Reinforcement Learning (202 KB) Robert Fitch, Bernhard Hengst, Dorian Suc, Greg Calbert, Jason Scholz, The 18th Australian Joint Conference on Artificial Intelligence, Sydney Australia December 2005.

Machine Learning for Adversarial Agent Microworlds (498 KB) J. Scholz, B. Hengst, G. Calbert, A. Antoniades, P. Smet, L. Marsh, H-W. Kwok, D. Gossink, MODSIM'05 Melbourne Australia December 2005.

2004

Model Approximation for HEXQ Hierarchical Reinforcement Learning (312 KB) Bernhard Hengst, 15th European Conference on Machine Learning (ECML), Pisa, Italy, © Springer-Verlag.

Concurrent Discovery of Task Hierarchies (570 KB) Duncan Potts, Bernhard Hengst,  AAAI Spring Symposium on Knowledge Representation and Ontology for Autonomous Systems 2004.

Discovering Multiple Levels of a Task Hierarchy Concurrently, Duncan Potts and Bernhard Hengst
Robotics and Autonomous Systems, Volume 49, Issues 1-2, 30 November 2004, Pages 43-55 PDF (273 K)

A Fast Vision Sensor Model: Matching Edges with NightOwl (399 KB) Raymond Sheh, Bernhard Hengst (2004) Proceedings of the Australasian Conference on Robotics and Automation (ACRA) http://robot.anu.edu.au/acra/ 

NightOwl: Self Localisation by Matching Edges (336 KB) Raymond Sheh, Bernhard Hengst (2004) UNSW-CSE-TR #0406, School of Computer Science and Engineering, University of New South Wales, Australia.

2003

Discovering Hierarchy in Reinforcement Learning, Bernhard Hengst, PhD Thesis, Computer Science and Engineering, University of New South Wales, Sydney Australia.

Variable Resolution in Hierarchical RL ( 88 KB) Bernhard Hengst (2003) UNSW-CSE-TR #0309, National ICT Australia, School of Computer Science and Engineering, University of New South Wales.

Safe State Abstraction and Discounting in Hierarchical Reinforcement Learning (209 KB) Bernhard Hengst (2003), UNSW-CSE-TR #0308, National ICT Australia, School of Computer Science and Engineering, University of New South Wales.

The UNSW RoboCup 2002 Legged League Team (344 KB)Andre Olave, David Wang, James Wong, TimothyTam, Benjamin Leung, Min Sub Kim, James Brooks, Albert Chang, Nik Von Huben, Claude Sammut, Bernhard Hengst (2003) The First RoboCup Australian Open 2003 (AORC-2003) 

2002

Discovering Hierarchy in Reinforcement Learning with HEXQ ( 305 KB)Hengst, B. (2002).  Proceedings of the Nineteenth International Conference on Machine Learning, edited by Claude Sammut and Achim Hoffmann, UNSW Sydney Australia ICML 2002, pp 243-250. This paper was also presented at the ICML 2002 workshop on Development of Representations edited by Edwin de Jong and Tim Oats.

Localisation in an Uncertain World with Multiple Agents (194 KB) Zheng Wang, Bernhard Hengst, Claude Sammut (2002)  MI19.

The Evolution of a Robot Soccer Team ( 91 KB) Claude Sammut and Bernhard Hengst (2002) . ISRR

The UNSW RoboCup 2001 Sony Legged League Team.  Spencer Chen, Martin Siu, Thomas Vogelgesang, Tak Fai Yik, Bernhard Hengst, Son Bao Pham, Claude Sammut (2001)  UNSW Thesis Report. Also in Lecture Notes in Computer Science, Lecture Notes in Artificial Intelligence LNAI 2377 A. Birk, S. Coradeschi, S. Tadokoro (Eds.): RoboCup 2001: Robot Soccer World Cup V, Springer 2002.

2001

The Fourth Robotic Soccer World Championships (5.3 MB) Edited by Peter Stone with Minoru Asada, Tucker Balch, Raffaelo D'Andrea, Masahiro Fujita, Bernhard Hengst, Gerhard Kraetzschmar, Pedro Lima, Nuno Lau, Henrik Lund, Daniel Polani, Paul Scerri, Satoshi Tadokoro, Thilo Weigel, and Gordon Wyeth. (2001) RoboCup-2000: . AI Magazine Volume 22, No1. Spring 2001.

The UNSW RoboCup 2000 Sony Legged League Team (158 KB) Hengst, B., Ibbotson, D., Pham, S.B., Dalgliesh, J., Lawther, M., Preston, P., Sammut, C. (2001) , in Peter Stone and Tucker Balch and Gerhard Kraetzschmar (Eds) RoboCup 2000: Robot Soccer World Cup IV, Lecture Notes in Artificial Intelligence, Vol 2019, pp64-75, Springer-Verlag, Berlin Heidelberg New York (copyright)

Learning Hierarchical Decomposition for Factored MDPs (51 KB) Hengst, B. (2001)  presented at ICML 2001 workshop on Hierarchy and Memory in Reinforcement Learning. See also ICML 2001 other workshop papers.

Omnidirectional Locomotion for Quadruped Robots (157 KB) Hengst, B., Ibbotson, D., Pham, S.B., P., Sammut, C. (2001)  RoboCup International Symposium, August 7-8, 2001 in Lecture Notes in Computer Science, Lecture Notes in Artificial Intelligence LNAI 2377 A. Birk, S. Coradeschi, S. Tadokoro (Eds.): RoboCup 2001: Robot Soccer World Cup V, Springer 2002. p.368 ff..

Stochastic Gradient Descent Localisation in Quadruped Robots ( 50KB) Son Bao Pham, Bernhard Hengst, Darren Ibbotson and Claude Sammut. , RoboCup International Symposium, August 7-8, 2001, in Lecture Notes in Computer Science, Lecture Notes in Artificial Intelligence LNAI 2377 A. Birk, S. Coradeschi, S. Tadokoro (Eds.): RoboCup 2001: Robot Soccer World Cup V, Springer 2002. p. 447 ff..

2000

Generating Hierarchical Structure in Reinforcement Learning from State Variables Hengst, B. , In Mizoguchi, R., Slaney, J. (Eds.) PRICAI 2000 Topics in Artificial Intelligence, (copyright) Springer

1978

Development and Implementation of a Hierarchical Computer Aided Company Planning System.  Hengst, B. In Rose, J. (ed.): Current Topics in Cybernetics and Systems, Proceedings of the Fourth International Congress of Cybernetics and Systems, Amsterdam, The Netherlands, Springer-Verlag, Berlin Heidelberg New York

1977

The Use of a Computer to Describe Blasting. Harries, G., Hengst, B., 15th APCOM Symposium, Brisbane, Australia