Biography
Professor Richard Dazeley is the Deputy Head of School for the School of Information Technology (SIT) at Deakin University. He is also the lead investigator in the School’s program of research into eXplainability for Human-aligned Autonomous Systems (XHAS) and the Leader for the Machine Intelligence Lab in the Centre for Data to Intelligence (D2I). Externally, he is a senior member of the International AI Existential Safety Community with the Future of Life Institute; founding and leading member of the International Multiobjective Reinforcement Learning community; and, the co-founder/leader of the Australian Responsible Autonomous Agents Collective (ARAAC). Previously, between 2018-2023, he had been the Associate Head of School (Research) and Course Director for the Master of Applied Artificial Intelligence and the Bachelor of Computer Science. At Federation University (2007-2018) he had been the Head of Discipline (Information Technology) and the Associate Dean Learning and Teaching.
Professor Dazeley is an internationally leading researcher in Multiobjective Reinforcement Learning (MORL) and its application to the Human-alignment problem. As one of the founding researchers of MORL he is an author of several of the key papers in the field; frequent internationally invited speaker; guest editor of journals; and workshop organiser in the field. His work in the human-alignment problem has resulted in his increasing profile in the debate on the emergence of Artificial General Intelligence (AGI); acceptance as a senior member of the prestigious international AI Existential safety community; invited to UC Berkeley’s NSF funded consortium on Safe and Beneficial AI; published in public forums; and, invitations as speaker to a range of international and national forums. He has led several major research projects as a chief investigator including international partnerships.
As an educator Professor Dazeley has been recognised as a nationally leading ICT Curriculum Designer – receiving the Australian National Award as ICT Educator of the Year (2016) from the Australian Computer Society (ACS) and was nominated for the International ICT Educator of the Year award from the South-East Asia Regional Computer Confederation (SEARCC) (2017). He has consulted closely with the development of the Australian Primary and Secondary National Curriculum, worked with the ACS and industry in the development of national industry standards, and provided consultation across a number of Australian universities on the development of innovative curriculum. He primarily teaches in the areas of machine learning, artificial intelligence, programming, algorithms, game/graphics programming.
Research interests
Reinforcement learning and the human-alignment problem - focusing on multi-objective, explainable, safe, ethical, interactive and multi-agent reinforcement learning methods. Also pursue interests in knowledge representation, acquisition and prudence analysis and their incorporation into machine learning methods.
Units taught
SIT796 - Reinforcement Learning
SIT799 - Human Aligned Artificial Intelligence
SIT307 - Data Mining and Machine Learning
SIT720 - Machine Learning
SIT221 - Data Structures and Algorithms
SIT102 - Introduction to Programming
Research groups
External: Senior member of the International AI Existential Safety Community with the Future of Life Institute
External: Co-leader of the Australian Responsible Autonomous Agents Collective (ARAAC).
External: Founding member of the International Multiobjective Reinforcement Learning Community (MORL-com).
Internal: Leader Machine Intelligence Lab (MIL).
Internal: Chief Investigator of the eXplainable Human-aligned Autonomous Systems (XHAS) Program of research.
Awards
2019 - Teaching and Learning Award
2017 - International ICT Educator of the Year award nominee from the South-East Asia Regional Computer Confederation (SEARCC)
2016 - ACS Australian National ICT Educator of the Year Award
2014 - Dean’s award for Outstanding Service
2011 - Vice-Chancellor’s Excellence in Service Award .
Publications
Elastic step DQN: A novel multi-step algorithm to alleviate overestimation in Deep Q-Networks
A Ly, R Dazeley, P Vamplew, F Cruz, S Aryal
(2024), Vol. 576, pp. 1-13, Neurocomputing, Amsterdam, The Netherlands, C1
A Wearable Multi-Sensor Fusion Approach for Gender Recognition based on Deep Learning
Supriya Roy, Bahareh Nakisa, Pubudu Pathirana, Richard Dazeley
(2024), pp. 114-119, ICBRA 2023 : Proceedings of the 10th International Conference on Bioinformatics Research and Applications 2023, Barcelona, Spain, E1
Overcoming weaknesses of density peak clustering using a data-dependent similarity measure
Z Rasool, S Aryal, M Bouadjenek, R Dazeley
(2023), Vol. 137, pp. 109287-109287, Pattern Recognition, C1
Explainable reinforcement learning for broad-XAI: a conceptual framework and survey
Richard Dazeley, Peter Vamplew, Francisco Cruz
(2023), Vol. 35, pp. 16893-16916, Neural Computing and Applications, Berlin, Germany, C1
H Nguyen, F Cruz, R Dazeley
(2023), Vol. 23, pp. 1-15, Sensors, Basel, Switzerland, C1
AI apology: interactive multi-objective reinforcement learning for human-aligned AI
H Harland, R Dazeley, B Nakisa, F Cruz, P Vamplew
(2023), Vol. 35, pp. 16917-16930, Neural Computing and Applications, Berlin, Germany, C1
A NetHack Learning Environment Language Wrapper for Autonomous Agents
N Goodger, P Vamplew, C Foale, R Dazeley
(2023), Vol. 11, pp. 1-10, Journal of Open Research Software, London, Eng., C1
Masked Autoencoders in 3D Point Cloud Representation Learning
Jincen Jiang, Xuequan Lu, Lizhi Zhao, Richard Dazaley, Meili Wang
(2023), pp. 1-12, IEEE Transactions on Multimedia, Piscataway, N.J., C1
Segmentation-driven feature-preserving mesh denoising
W Wang, W Pan, C Dai, R Dazeley, L Wei, B Rolfe, X Lu
(2023), pp. 1-17, Visual Computer, Berlin, Germany, C1
Peter Vamplew, Benjamin Smith, Johan Källström, Gabriel Ramos, Roxana Rădulescu, Diederik Roijers, Conor Hayes, Friedrik Heintz, Patrick Mannion, Pieter Libin, Richard Dazeley, Cameron Foale
(2023), AAMAS-2023 : Proceedings of the 22nd International Conference on Autonomous Agents and Multiagent Systems 2023, London, Eng., E1
A Brief Guide to Multi-Objective Reinforcement Learning and Planning
Conor Hayes, Roxana Rădulescu, Eugenio Bargiacchi, Johan Källström, Matthew Macfarlane, Mathieu Reymond, Timothy Verstraeten, Luisa Zintgraf, Richard Dazeley, Fredrik Heintz, Enda Howley, Athirai Irissappane, Patrick Mannion, Ann Nowé, Gabriel Ramos, Marcello Restelli, Peter Vamplew, Diederik Roijers
(2023), pp. 1988-1990, AAMAS-2023 : Proceedings of the 22nd International Conference on Autonomous Agents and Multiagent Systems 2023, London, Eng., E1
Elastic step DDPG: Multi-step reinforcement learning for improved sample efficiency
A Ly, R Dazeley, P Vamplew, F Cruz, S Aryal
(2023), Vol. 2023-June, pp. 1-6, IJCNN 2023 : Proceedings of the International Joint Conference on Neural Networks, Gold Coast, Queensland, E1
Weighted Point Cloud Normal Estimation
W Wang, X Lu, D Shao, X Liu, R Dazeley, A Robles-Kelly, W Pan
(2023), Vol. 2023-July, pp. 2015-2020, ICME 2023 : Proceedings of the IEEE International Conference on Multimedia and Expo, Brisbane, Queensland, E1
The impact of environmental stochasticity on value-based multiobjective reinforcement learning
P Vamplew, C Foale, R Dazeley
(2022), Vol. 34, pp. 1783-1799, Neural Computing and Applications, C1
Discrete-to-deep reinforcement learning methods
B Kurniawan, P Vamplew, M Papasimeon, R Dazeley, C Foale
(2022), Vol. 34, pp. 1713-1733, Neural Computing and Applications, Berlin, Germany, C1
Human engagement providing evaluative and informative advice for interactive reinforcement learning
A Bignold, F Cruz, R Dazeley, P Vamplew, C Foale
(2022), pp. 1-16, Neural Computing and Applications, Berlin, Germany, C1
T Chuong, V Mak-Hau, J Yearwood, R Dazeley, M Nguyen, T Cao
(2022), Vol. 319, pp. 1533-1564, Annals of Operations Research, New York, N.Y., C1
A practical guide to multi-objective reinforcement learning and planning
C Hayes, R Rădulescu, E Bargiacchi, J Källström, M Macfarlane, M Reymond, T Verstraeten, L Zintgraf, R Dazeley, F Heintz, E Howley, A Irissappane, P Mannion, A Nowé, G Ramos, M Restelli, P Vamplew, D Roijers
(2022), Vol. 36, pp. 1-59, Autonomous Agents and Multi-Agent Systems, Berlin, Germany, C1
Scalar reward is not enough: a response to Silver, Singh, Precup and Sutton (2021)
P Vamplew, B Smith, J Källström, G Ramos, R Rădulescu, D Roijers, C Hayes, F Heintz, P Mannion, P Libin, R Dazeley, C Foale
(2022), Vol. 36, Autonomous Agents and Multi-Agent Systems, C1
Evaluating Human-like Explanations for Robot Actions in Reinforcement Learning Scenarios
F Cruz, C Young, R Dazeley, P Vamplew
(2022), pp. 894-901, IROS 2022 : Proceedings of the IEEE International Conference on Intelligent Robots and Systems 2022, Kyoto, Japan, E1
usfAD: a robust anomaly detector based on unsupervised stochastic forest
S Aryal, K Santosh, R Dazeley
(2021), Vol. 12, pp. 1137-1150, International Journal of Machine Learning and Cybernetics, C1
Potential-based multiobjective reinforcement learning approaches to low-impact agents for AI safety
P Vamplew, C Foale, R Dazeley, A Bignold
(2021), Vol. 100, Engineering Applications of Artificial Intelligence, C1
An evaluation methodology for interactive reinforcement learning with simulated users
A Bignold, F Cruz, R Dazeley, P Vamplew, C Foale
(2021), Vol. 6, pp. 1-15, Biomimetics, Switzerland, C1
A Prioritized objective actor-critic method for deep reinforcement learning
N Nguyen, T Nguyen, P Vamplew, R Dazeley, S Nahavandi
(2021), Vol. 33, pp. 10335-10349, Neural Computing and Applications, C1
Levels of explainable artificial intelligence for human-aligned conversational explanations
R Dazeley, P Vamplew, C Foale, C Young, S Aryal, F Cruz
(2021), Vol. 299, Artificial Intelligence, C1
A Robust Approach for Continuous Interactive Actor-Critic Algorithms
C Millan-Arias, B Fernandes, F Cruz, R Dazeley, S Fernandes
(2021), Vol. 9, pp. 104242-104260, IEEE Access, C1
Explainable robotic systems: understanding goal-driven actions in a reinforcement learning scenario
F Cruz, R Dazeley, P Vamplew, I Moreira
(2021), Neural Computing and Applications, C1
Persistent rule-based interactive reinforcement learning
A Bignold, F Cruz, R Dazeley, P Vamplew, C Foale
(2021), pp. 1-18, Neural Computing and Applications, Berlin, Germany, C1
A conceptual framework for externally-influenced agents: an assisted reinforcement learning review
A Bignold, F Cruz, M Taylor, T Brys, R Dazeley, P Vamplew, C Foale
(2021), Journal of Ambient Intelligence and Humanized Computing, Berlin, Germany, C1
Language Representations for Generalization in Reinforcement Learning
N Goodger, P Vamplew, C Foale, R Dazeley
(2021), Vol. 157, pp. 390-405, ACML 2021 : Proceedings of the 13th Asian Conference on Machine Learning 2021, Virtual, E1
A multi-objective deep reinforcement learning framework
Thanh Nguyen, Ngoc Nguyen, Peter Vamplew, Saeid Nahavandi, Richard Dazeley, Chee Lim
(2020), Vol. 96, pp. 1-12, Engineering Applications of Artificial Intelligence, Amsterdam, The Netherlands, C1
Deep reinforcement learning with interactive feedback in a human-robot environment
I Moreira, J Rivas, F Cruz, R Dazeley, A Ayala, B Fernandes
(2020), Vol. 10, Applied Sciences (Switzerland), C1
A comparison of humanoid robot simulators: a quantitative approach
A Ayala, F Cruz Naranjo, D Campos, R Rubio, B Fernandes, R Dazeley
(2020), pp. 1-6, ICDL-EpiRob 2020 : Proceedings of the 10th IEEE International Conference on Development and Learning and Epigenetic Robotics, Online from Valparaiso, Chile, E1
B Kurniawan, P Vamplew, M Papasimeon, R Dazeley, C Foale
(2019), Vol. 11919, pp. 54-65, AI 2019: Advances in artificial intelligence : Proceedings of the 32nd Australian Joint Conference on Artificial Intelligence 2019, Adelaide, S. Aust., E1
Memory-based explainable reinforcement learning
F Cruz Naranjo, R Dazeley, P Vamplew
(2019), Vol. 11919, pp. 66-77, AI 2019: Advances in artificial intelligence : Proceedings of the 32nd Australasian Joint Conference on Artificial Intelligence 2019, Adelaide, S. Aust., E1
Aggregation of dependent criteria in multicriteria decision making problems by means of capacities
G Beliakov, S Boswell, T Cao, R Dazeley, V Mak-Hau, M Nguyen, T Wilkin, J Yearwood
(2019), pp. 228-234, MODSIM2019 : Supporting evidence-based decision making: the role of modelling and simulation : Proceedings of the 23rd International Congress on Modelling and Simulation, Canberra, A.C.T., E1
Human-aligned artificial intelligence is a multiobjective problem
P Vamplew, R Dazeley, C Foale, S Firmin, J Mummery
(2018), Vol. 20, pp. 27-40, Ethics and information technology, New York, N.Y., C1-1
Non-functional regression: a new challenge for neural networks
P Vamplew, R Dazeley, C Foale, T Choudhury
(2018), Vol. 314, pp. 326-335, Neurocomputing, Amsterdam, The Netherlands, C1
Britt Klein, Lisa Clinnick, Jessica Chesler, Andrew Stranieri, Adam Bignold, Richard Dazeley, Suzanne McLaren, Sue Lauder, Venki Balasubramanian
(2018), Vol. 246, pp. 24-28, Telehealth for our ageing society : Proceedings of the 5th Global Telehealth Meeting 2017, Adelaide, S. Aust., E1-1
Rapid anomaly detection using integrated prudence analysis (IPA)
O Maruatona, P Vamplew, R Dazeley, P Watters
(2018), Vol. 11154, pp. 137-141, PAKDD 2018 : Trends and applications in knowledge discovery and data mining : the 22nd Pacific-Asia Conference on Knowledge Discovery and Data Mining, Melbourne, Victoria, E1-1
Steering approaches to Pareto-optimal multiobjective reinforcement learning
P Vamplew, R Issabekov, R Dazeley, C Foale, A Berry, T Moore, D Creighton
(2017), Vol. 263, pp. 26-38, Neurocomputing, Amsterdam, The Netherlands, C1-1
Softmax exploration strategies for multiobjective reinforcement learning
P Vamplew, R Dazeley, C Foale
(2017), Vol. 263, pp. 74-86, Neurocomputing, Amsterdam, The Netherlands, C1-1
Evaluating accuracy in prudence analysis for cyber security
O Maruatona, P Vamplew, R Dazeley, P Watters
(2017), Vol. 10638, pp. 407-417, ICONIP 2017 : Proceedings of the Neural Information Processing International Conference, Guangzhou, China, E1-1
Authorship analysis of aliases: Does topic influence accuracy?
R Layton, P Watters, R Dazeley
(2015), Vol. 21, pp. 497-518, Natural language engineering, London, Eng., C1-1
Reinforcement learning of pareto-optimal multiobjective policies using steering
P Vamplew, R Issabekov, R Dazeley, C Foale
(2015), Vol. 9457, pp. 596-608, AI 2015 : Proceedings of the 28th Australasian Joint Conference on Artificial Intelligence 2015, Canberra, A.C.T., E1-1
A survey of multi-objective sequential decision-making
D Roijers, P Vamplew, S Whiteson, R Dazeley
(2013), Vol. 48, pp. 67-113, Journal of artificial intelligence research, Palo Alto, Calif., C1-1
Evaluating authorship distance methods using the positive silhouette coefficient
R Layton, P Watters, R Dazeley
(2013), Vol. 19, pp. 517-535, Natural language engineering, Cambridge, Eng., C1-1
Automated unsupervised authorship analysis using evidence accumulation clustering
R Layton, P Watters, R Dazeley
(2013), Vol. 19, pp. 95-120, Natural language engineering, Cambridge, Eng., C1-1
A survey of multi-objective sequential decision-making
Diederik Roijers, Peter Vamplew, Shimon Whiteson, Richard Dazeley
(2013), Vol. 48, Journal of artificial intelligence research, C1-1
Local n-grams for author identification: notebook for PAN at CLEF 2013
R Layton, P Watters, R Dazeley
(2013), Vol. 1179, pp. 1-4, CLEF 2013 : Proceedings of the CLEF 2013 Conference, Valencia, Spain, E1-1
Recentred local profiles for authorship attribution
R Layton, P Watters, R Dazeley
(2012), Vol. 18, pp. 293-312, Natural language engineering, Cambridge, Eng., C1-1
Detection of CAN by ensemble classifiers based on ripple down rules
A Kelarev, R Dazeley, A Stranieri, J Yearwood, H Jelinek
(2012), Vol. 7457, pp. 147-159, PKAW 2012 : Proceedings of the 12th International Workshop on Knowledge Management and Aquisition for Intelligent Systems, Kuching, Malaysia, E1-1
Prudent fraud detection in internet banking
O Maruatona, P Vamplew, R Dazeley
(2012), pp. 60-65, Proceedings - 2012 3rd Cybercrime and Trustworthy Computing Workshop, CTC 2012, AUSTRALIA, Univ Ballarat, Ballarat, E1-1
RM and RDM, a preliminary evaluation of two prudent RDR techniques
O Maruatona, P Vamplew, R Dazeley
(2012), Vol. 7457 LNAI, pp. 188-194, PKAW 2012 : Proceedings of the 12th International Workshop on Knowledge Management and Acquisition for Intelligent Systems, Kuching, Malaysia, E1-1
Unsupervised authorship analysis of phishing webpages
R Layton, P Watters, R Dazeley
(2012), pp. 1104-1109, 2012 International Symposium on Communications and Information Technologies, ISCIT 2012, Gold Coast, QLD, E1-1
Establishing reasoning communities of security experts for internet commerce security
A Kelarev, S Brown, P Watters, X Wu, R Dazeley
(2011), pp. 380-396, Technologies for supporting reasoning communities and collaborative decision making: Cooperative approaches, Hershey, Pa., B1-1
Online knowledge validation with prudence analysis in a document management application
R Dazeley, S Park, B Kang
(2011), Vol. 38, pp. 10959-10965, Expert systems with applications, Amsterdam, The Netherlands, C1-1
Empirical evaluation methods for multiobjective reinforcement learning algorithms
P Vamplew, R Dazeley, A Berry, R Issabekov, E Dekker
(2011), Vol. 84, pp. 51-80, Machine learning, New York, N.Y., C1-1
How much material on BitTorrent is infringing content? A case study
P Watters, R Layton, R Dazeley
(2011), Vol. 16, pp. 79-87, Information security technical report, Amsterdam, The Netherlands, C1-1
Authorship attribution for Twitter in 140 characters or less
R Layton, P Watters, R Dazeley
(2010), pp. 1-8, CTC 2010 : Proceedings of the 2nd Cybercrime and Trustworthy Computing Workshop, Ballarat, VIC, E1-1
The Ballarat incremental knowledge engine
R Dazeley, P Warner, S Johnson, P Vamplew
(2010), Vol. 6232, pp. 195-207, PKAW 2010 : Proceedings of the Pacific Rim Knowledge Acquisition Workshop : Knowledge Management and Acquisition for Smart Systems and Services, Daegu, South Korea, E1-1
Optimization of multiple classifiers in data mining based on string rewriting systems
R Dazeley, A Kelarev, J Yearwood, M Mammadov
(2009), Vol. 2, pp. 41-56, Asian-European journal of mathematics, C1-1
Generalising symbolic knowledge in online classification and prediction
R Dazeley, B Kang
(2009), Vol. 5465, pp. 91-108, Knowledge Acquisition: Approaches, Algorithms and Applications: Pacific Rim Knowledge Acquisition Workshop, PKAW 2008, Hanoi, Vietnam, December 15-16, 2008, Revised Selected Papers, Hanoi, Vietnam, E1-1
Constructing stochastic mixture policies for episodic multiobjective reinforcement learning tasks
P Vamplew, R Dazeley, E Barker, A Kelarev
(2009), Vol. 5866, pp. 340-349, AI 2009 : Advances in Artificial Intelligence : Proceedings of the 22nd Australasian Joint Conference, Melbourne, Vic., E1-1
Epistemological approach to the process of practice
R Dazeley, B Kang
(2008), Vol. 18, pp. 547-567, Minds and machines, Dordrecht, The Netherlands, C1-1
An expert system methodology for SMEs and NPOs
R Dazeley
(2008), Harnessing Knowledge Management to Build Communities - Proceedings of the 11th Annual Australian Conference on Knowledge Management and Intelligent Decision Support, ACKMIDS 08, Mount Helen, Vic., E1-1
Detecting the knowledge boundary with prudence analysis
R Dazeley, B Kang
(2008), Vol. 5360 LNAI, pp. 482-488, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Auckland Univ Technol, Auckland, NEW ZEALAND, E1-1
An approach for generalising symbolic knowledge
R Dazeley, B Kang
(2008), Vol. 5360 LNAI, pp. 379-385, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Auckland Univ Technol, Auckland, NEW ZEALAND, E1-1
Weighted MCRDR: Deriving information about relationships between classifications in MCRDR
R Dazeley, B Kang
(2003), Vol. 2903, pp. 245-255, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), UNIV WESTERN AUSTRALIA, PERTH, AUSTRALIA, E1-1
Funded Projects at Deakin
Other Public Sector Funding
Scaling Force Effectiveness Modelling and Assessment Methods.
Prof Vicky Mak, Prof John Yearwood, Prof Gleb Beliakov, Prof Richard Dazeley, A/Prof Tim Wilkin
The Defence Science and Technology Group of the Department of Defence
- 2021: $49,221
- 2020: $114,880
- 2019: $131,293
Modeling Adversary Intent Using Multiobjective Reinforcement Learning.
Prof Richard Dazeley, Dr Sunil Aryal
DSTO Grant - Research - Defence Science & Technology Organisation
- 2021: $53,728
A competency-aware multi-agent framework for human-machine teams in adversarial environments.
Prof Richard Dazeley, Dr Sunil Aryal, A/Prof Tim Wilkin
DSTO Grant - Research - Defence Science & Technology Organisation
- 2022: $26,118
- 2021: $125,124
Application of Generic Actual Argument Model to represent complex decisions and generate narratives.
Prof Richard Dazeley, Dr Sunil Aryal, Dr Bahadorreza Ofoghi, Prof John Yearwood
Department of Defence
- 2022: $30,000
AI Apology: Considerations for An Apologetic Approach to Socially Responsible Agents.
Prof Richard Dazeley, Dr Bahareh Nakisa, Mx Hadassah Harland
CSIRO
- 2024: $10,000
- 2023: $10,000
Industry and Other Funding
Empowering Agents with Human Feedback: Integrating Implicit and Explicit Signals in Deep Reinforcement Learning
Dr Bahareh Nakisa, Dr Naim Rastgoo, Prof Richard Dazeley, Dr Sunil Aryal
Asian Office of Aerospace Research and Development
- 2023: $256,742
Intelligent Table Tennis Match Analysis: Automated Insights Extraction through Machine Learning and Computer Vision.
Dr Sunil Aryal, Dr Bahareh Nakisa, Prof Richard Dazeley
Table Tennis Australia Ltd
- 2023: $20,000
Supervisions
No completed student supervisions to report