Detecting the knowledge boundary with prudence analysis
- Authors: Dazeley, Richard , Kang, Byeongho
- Date: 2008
- Type: Text , Conference paper
- Relation: Paper presented at 21st Australasian Joint Conference on Artificial Intelligence, Auckland, New Zealand : 1st-5th December 2008 p. 482-488
- Full Text: false
- Description: Prudence analysis (PA) is a relatively new, practical and highly innovative approach to solving the problem of brittleness in knowledge based systems (KBS). PA is essentially an online validation approach, where as each situation or case is presented to the KBS for inferencing the result is simultaneously validated. This paper introduces a new approach to PA that analyses the structure of knowledge rather than the comparing cases with archived situations. This new approach is positively compared against earlier systems for PA, strongly indicating the viability of the approach.
- Description: 2003006511
Grid-based information retrieval for the aggregation of legal datasets in online dispute resolution
- Authors: Saeed, Ather , Stranieri, Andrew , Dazeley, Richard , Ma, Liping
- Date: 2009
- Type: Text , Journal article
- Relation: Communications of SIWN Vol. 6, no. April (2009), p. 16-22
- Full Text: false
- Description: The Web is a stateless and complex environment when it comes to the retrieval of information from millions of computers connected to the Internet via WWW servers. Information Retrieval (IR) from heterogeneous data sources poses a great challenge as the information of interest is stored in a variety of different formats. Answering an enormous amount of queries is a resource and computational intensive task in ODR (Online Dispute Resolution). Information availability also poses a challenge when it comes to the mediation and arbitration processes in resolving eCommerce and legal disputes. A new Grid-based information retrieval model is proposed for the aggregation and replication of legal datasets from remote machines with indexed-based search facility. Datasets of interests will be indexed with a slight modification to the existing indexing scheme. A new strategy is proposed to deal with similar queries posted over and over again and how the commonality among the XML query trees are exploited and merged for the efficient retrieval of information.
Online knowledge validation with prudence analysis in a document management application
- Authors: Dazeley, Richard , Park, Sung Sik , Kang, Byeongho
- Date: 2011
- Type: Text , Journal article
- Relation: Expert Systems with Applications Vol. , no. (2011), p.
- Full Text: false
- Reviewed:
- Description: Prudence analysis (PA) is a relatively new, practical and highly innovative approach to solving the problem of brittleness in knowledge based system (KBS) development. PA is essentially an online validation approach where as each situation or case is presented to the KBS for inferencing the result is simultaneously validated. Therefore, instead of the system simply providing a conclusion, it also provides a warning when the validation fails. Previous studies have shown that a modification to multiple classification ripple-down rules (MCRDR) referred to as rated MCRDR (RM) has been able to achieve strong and flexible results in simulated domains with artificial data sets. This paper presents a study into the effectiveness of RM in an eHealth document monitoring and classification domain using human expertise. Additionally, this paper also investigates what affect PA has when the KBS developer relied entirely on the warnings for maintenance. Results indicate that the system is surprisingly robust even when warning accuracy is allowed to drop quite low. This study of a previously little touched area provides a strong indication of the potential for future knowledge based system development. © 2011 Elsevier Ltd. All rights reserved.
On the limitations of scalarisation for multi-objective reinforcement learning of Pareto fronts
- Authors: Vamplew, Peter , Yearwood, John , Dazeley, Richard , Berry, Adam
- Date: 2008
- Type: Text , Conference paper
- Relation: Paper presented at 21st Australasian Joint Conference on Artificial Intelligence, Auckland, New Zealand : 1st-5th December 2008 Vol. 5360, p. 372-378
- Full Text: false
- Description: Multiobjective reinforcement learning (MORL) extends RL to problems with multiple conflicting objectives. This paper argues for designing MORL systems to produce a set of solutions approximating the Pareto front, and shows that the common MORL technique of scalarisation has fundamental limitations when used to find Pareto-optimal policies. The work is supported by the presentation of three new MORL benchmarks with known Pareto fronts.
- Description: 2003006504
Scalable continuous query architecture for eCommerce and legal disputes
- Authors: Saeed, Ather , Stranieri, Andrew , Dazeley, Richard , Ma, Liping
- Date: 2008
- Type: Text , Journal article
- Relation: Communications of SIWN Vol. 3, no. (2008), p. 1-6
- Full Text: false
- Reviewed:
- Description: Continuous Queries (CQ) are persistent, content sensitive and time dependent. Once the CQ is installed it will continuously poll the data sources and monitor updates of interest. This paper discusses major problems and issues with the existing CQ techniques for monitoring updates of interest on the web. A new Continuous Query based architecture is proposed to deal with the context sensitive problems of negotiation, mediation and arbitration to resolve Ecommerce and legal disputes. A business process model is given to automate mediation and arbitration processes in ODR (Online dispute resolution) to resolve disputes efficiently and in a timely manner. In the proposed CQ-Mediator architecture partial page update and web services are integrated for efficient monitoring and notification of updates to the disputants, mediators and arbitrators. Performance results of the proposed architecture and business process model for CQ-based ODR is also discussed in the experiment section.
- Description: 2003006852
An approach for generalising symbolic knowledge
- Authors: Dazeley, Richard , Kang, Byeongho
- Date: 2008
- Type: Text , Conference paper
- Relation: Paper presented at 21st Australasian Joint Conference on Artificial Intelligence, Auckland, New Zealand : 1st-5th December 2008 p. 379-385
- Full Text: false
- Description: Many researchers and developers of knowledge based systems (KBS) have been incorporating the notion of context. However, they generally treat context as a static entity, neglecting many connectionists’ work in learning hidden and dynamic contexts, which aids generalization. This paper presents a method that models hidden context within a symbolic domain achieving a level of generalisation. Results indicate that the method can learn the information that experts have difficulty providing by generalising the captured knowledge.
- Description: 2003006525
How much material on BitTorrent is infringing content? A case study
- Authors: Watters, Paul , Layton, Robert , Dazeley, Richard
- Date: 2011
- Type: Text , Journal article
- Relation: Information Security Technical Report Vol. 16, no. 2 (2011), p. 79-87
- Full Text: false
- Reviewed:
- Description: BitTorrent is a widely used protocol for peer-to-peer (P2P) file sharing, including material which is often suspected to be infringing content. However, little systematic research has been undertaken to establish to measure the true extent of illegal file sharing. In this paper, we propose a new methodology for measuring the extent of infringing content. Our initial results indicate that at least 89.9% of files shared contain infringing content, with a replication study on another sample finding 97%. We discuss the limitations of the approach in this case study, including sampling biases, and outline proposals to further verify the results. The implications of the work vis - vis the management of piracy at the network level are discussed. © 2011 Published by Elsevier Ltd. All rights reserved.
Empirical evaluation methods for multiobjective reinforcement learning algorithms
- Authors: Vamplew, Peter , Dazeley, Richard , Berry, Adam , Issabekov, Rustam , Dekker, Evan
- Date: 2011
- Type: Text , Journal article
- Relation: Machine Learning Vol. 84, no. 1-2 (2011), p. 51-80
- Full Text: false
- Reviewed:
- Description: While a number of algorithms for multiobjective reinforcement learning have been proposed, and a small number of applications developed, there has been very little rigorous empirical evaluation of the performance and limitations of these algorithms. This paper proposes standard methods for such empirical evaluation, to act as a foundation for future comparative studies. Two classes of multiobjective reinforcement learning algorithms are identified, and appropriate evaluation metrics and methodologies are proposed for each class. A suite of benchmark problems with known Pareto fronts is described, and future extensions and implementations of this benchmark suite are discussed. The utility of the proposed evaluation methods are demonstrated via an empirical comparison of two example learning algorithms. © 2010 The Author(s).
Detection of CAN by ensemble classifiers based on Ripple Down rules
- Authors: Kelarev, Andrei , Dazeley, Richard , Stranieri, Andrew , Yearwood, John , Jelinek, Herbert
- Date: 2012
- Type: Text , Book chapter
- Relation: Knowledge Management and Acquisition for Intelligent Systems p. 147-159
- Full Text: false
- Reviewed:
- Description: It is well known that classification models produced by the Ripple Down Rules are easier to maintain and update. They are compact and can provide an explanation of their reasoning making them easy to understand for medical practitioners. This article is devoted to an empirical investigation and comparison of several ensemble methods based on Ripple Down Rules in a novel application for the detection of cardiovascular autonomic neuropathy (CAN) from an extensive data set collected by the Diabetes Complications Screening Research Initiative at Charles Sturt University. Our experiments included essential ensemble methods, several more recent state-of-the-art techniques, and a novel consensus function based on graph partitioning. The results show that our novel application of Ripple Down Rules in ensemble classifiers for the detection of CAN achieved better performance parameters compared with the outcomes obtained previously in the literature.
Generalisation with symbolic knowledge in online classification
- Authors: Kang, Byeongho , Dazeley, Richard
- Date: 2008
- Type: Text , Conference paper
- Relation: PKAW-08: Proceedings of the Pacific Rim Knowledge Acquisition Workshop 2008
- Full Text: false
- Reviewed:
- Description: Increasingly, researchers and developers of knowledge based systems (KBS) have been incorporating the notion of context. For instance, Repertory Grids, Formal Concept Analysis (FCA) and Ripple-Down Rules (RDR) all integrate either implicit or explicit contextual information. However, these methodologies treat context as a static entity, neglecting many connectionists’ work in learning hidden and dynamic contexts, which aid their ability to generalize. This paper presents a method that models hidden context within a symbolic domain in order to achieve a level of generalisation. The method developed builds on the already established Multiple Classification Ripple-Down Rules (MCRDR) approach and is referred to as Rated MCRDR (RM). RM retains a symbolic core, while using a connection based approach to learn a deeper understanding of the captured knowledge. This method is applied to a number of online classification environments and results indicate that the method can learn the information that experts have difficulty providing.
RM and RDM, a preliminary evaluation of two prudent RDR Techniques
- Authors: Maruatona, Omaru , Vamplew, Peter , Dazeley, Richard
- Date: 2012
- Type: Text , Book chapter
- Relation: Knowledge Management and acquisition for intelligent systems: 12th Pacific Rim Knowledge Acquisition workshop p. 188-194
- Full Text: false
- Reviewed:
- Description: Rated Multiple Classification Ripple Down Rules (RM) and Ripple Down Models (RDM) are two of the successful prudent RDR approaches published. To date, there has not been a published, dedicated comparison of the two. This paper presents a systematic preliminary evaluation and analysis of the two techniques. The tests and results reported in this paper are the first phase of direct evaluations of RM and RDM against each other.
Establishing reasoning communities of security experts for Internet Commerce Security
- Authors: Kelarev, Andrei , Brown, Simon , Watters, Paul , Wu, Xinwen , Dazeley, Richard
- Date: 2011
- Type: Text , Book chapter
- Relation: Technologies for supporting reasoning communities and collaborative decision making : Cooperative approaches p. 380-396
- Full Text: false
- Reviewed:
- Description: The highly sophisticated and rapidly evolving area of internet commerce security presents many novel challenges for the organization of discourse in reasoning communities. This chapter suggests appropriate reasoning methods and demonstrates how establishing reasoning communities of security experts and enabling productive group discourse among them can play a crucial role in successful resolution of problems concerning the implementation, integration, deployment and maintenance of flexible local security systems for defense against malware threats in internet security. Local security systems of this sort may combine several ready open source or commercial software packages behind a common front-end and may enhance and supplement their facilities with additional plug-ins. To illustrate the diverse character of challenges the reasoning communities in internet security are likely to be faced with, this chapter concentrates on defense against phishing attacks. This example was selected as it is one of the newest and most rapidly changing application domains for the principles of organizing reasoning communities. The major group discourse methods suggested for the reasoning communities of security experts in this chapter include the Delphi Method, the Wideband Delphi Process, the Generic/Actual Argument Model of Structured Reasoning, Brainstorming, Reverse Brainstorming, Consensus Decision Making, Voting, Open Delphi and Open Brainstorming Methods. The Delphi Method and Wideband Delphi Process are suggested as tools for organizing a cohesive reasoning architecture, for coordinating other methods, and for preparing and allocating other methods to particular issues.
Energy-efficient priority-based routing scheme for the healthcare wireless sensor networks
- Authors: Saeed, Ather , Stranieri, Andrew , Dazeley, Richard
- Date: 2014
- Type: Text , Conference paper
- Relation: 9th WSEAS International Conference on Remote Sensing, Budapest 10/12/2103 pg 19-27
- Full Text: false
- Reviewed:
- Description: Abstract: - In time-critical and data intensive applications, efficient acquisition of sensitive datasets is a challenge because of network congestion, void regions and node failures that commonly occur in wireless sensor networks (WSN), while monitoring the wellbeing of patients with serious medical conditions. The sensor devices attached to such patients are used for monitoring the vital signs of those with serious heart problems, Parkinson disease, Epilepsy and high blood pressure. This paper typically focuses on the reliable acquisition of datasets and provides a fault-tolerant priority based routing scheme with Dynamic Jumping (FTMPR-DJ) for the energy-efficient acquisition and dissemination of datasets. A new fault-tolerant scheme has been proposed that will significantly minimize data loss and network congestion and is well supported with extensive experiments to show effectiveness of the proposed routing scheme.
Coarse Q-Learning : Addressing the convergence problem when quantizing continuous state variables
- Authors: Dazeley, Richard , Vamplew, Peter , Bignold, Adam
- Date: 2015
- Type: Text , Conference paper
- Relation: 2nd Multidisciplinary Conference on Reinforcement Learning and Decision Making
- Full Text: false
- Reviewed:
- Description: Value-based approaches to reinforcement learning (RL) maintain a value function that measures the long term utility of a state or state-action pair. A long standing issue in RL is how to create a finite representation in a continuous, and therefore infinite, state environment. The common approach is to use function approximators such as tile coding, memory or instance based methods. These provide some balance between generalisation, resolution, and storage, but converge slowly in multidimensional state environments. Another approach of quantizing state into lookup tables has been commonly regarded as highly problematic, due to large memory requirements and poor generalisation. In particular , attempting to reduce memory requirements and increase generalisation by using coarser quantization forms a non-Markovian system that does not converge. This paper investigates the problem in using quantized lookup tables and presents an extension to the Q-Learning algorithm, referred to as Coarse Q-Learning (C QL), which resolves these issues. The presented algorithm will be shown to drastically reduce the memory requirements and increase generalisation by simulating the Markov property. In particular, this algorithm means the size of the input space is determined by the granularity required by the policy being learnt, rather than by the inadequacies of the learning algorithm or the nature of the state-reward dynamics of the environment. Importantly, the method presented solves the problem represented by the curse of dimensionality.
Real-time self-stabilizing scheme for the localization of faults in wireless sensor networks
- Authors: Saeed, Ather , Stranieri, Andrew , Dazeley, Richard
- Date: 2014
- Type: Text , Book chapter
- Relation: Recents advances in Image, Audio and Signal Processing p. 233-242
- Full Text: false
- Reviewed:
- Description: Reliable acquisition of data from massively dense wireless sensor networks (WSN) is a challenge due to the unpredictable behaviour of nodes responsible for collecting and disseminating datasets of interest. Therefore, accurate sensing of events from nodes depend on several microscopic and macroscopic factors such as distance of a node from the sink, radio signal strength and connectedness of network for routing datasets to the nearest sink. Several Clustering schemes have been proposed for routing datasets, where major focus was on finding the next cluster-head with maximum energy for routing data. Such schemes are not suitable for the real-time dissemination of datasets because electing the next cluster-head is a computational intensive process. A new energy-efficient self-stabilizing sliding rectangle protocol (ESSRP) is proposed in this paper for ensuring reliability and connectedness of regions for minimizing data loss and prolonging network life. The proposed scheme not only looks at the energy-balance of a particular cluster but also ensures fault-localization and tolerance by providing self-stabilization to network in the event of nodes or links failure using Green’s Theorem. The WSN rectangular regions should be oriented counter-clockwise, piecewise regular and continuously differentiable so that faults can be efficiently localized, identified and rectified in a particular region
Reinforcement learning of pareto-optimal multiobjective policies using steering
- Authors: Vamplew, Peter , Issabekov, Rustam , Dazeley, Richard , Foale, Cameron
- Date: 2015
- Type: Text , Conference paper
- Relation: 28th Australasian Joint Conference on Artificial Intelligence, AI 2015; Canberra, ACT; 30th November-4th December 2015 Vol. 9457, p. 596-608
- Full Text: false
- Reviewed:
- Description: There has been little research into multiobjective reinforcement learning (MORL) algorithms using stochastic or non-stationary policies, even though such policies may Pareto-dominate deterministic stationary policies. One approach is steering which forms a nonstationary combination of deterministic stationary base policies. This paper presents two new steering algorithms designed for the task of learning Pareto-optimal policies. The first algorithm (w-steering) is a direct adaptation of previous approaches to steering, and therefore requires prior knowledge of recurrent states which are guaranteed to be revisited. The second algorithm (Q-steering) eliminates this requirement. Empirical results show that both algorithms perform well when given knowledge of recurrent states, but that Q-steering provides substantial performance improvements over w-steering when this knowledge is not available. © Springer International Publishing Switzerland 2015.
Rapid anomaly detection using integrated prudence analysis (IPA)
- Authors: Maruatona, Omaru , Vamplew, Peter , Dazeley, Richard , Watters, Paul
- Date: 2018
- Type: Text , Conference proceedings
- Relation: PAKDD 2018.Trends and Applications in Knowledge Discovery and Data Mining. p. 137-141
- Full Text: false
- Reviewed:
- Description: Integrated Prudence Analysis has been proposed as a method to maximize the accuracy of rule based systems. The paper presents evaluation results of the three Prudence methods on public datasets which demonstrate that combining attribute-based and structural Prudence produces a net improvement in Prudence Accuracy.
Supporting regional aged care nursing staff to manage residents’ behavioural and psychological symptoms of dementia, in real time, using the nurses’ behavioural assistant (NBA) : A pilot site 'end-user attitudes’ trial
- Authors: Klein, Britt , Clinnick, Lisa , Chesler, Jessica , Stranieri, Andrew , Bignold, Adam , Dazeley, Richard , McLaren, Suzanne , Lauder, Sue , Balasubramanian, Venki
- Date: 2018
- Type: Text , Conference paper
- Relation: 2017 Global Telehealth Meeting, GT 201; Adelaide, Australia; 22nd-24th November 2017; published in Telehealth for our Ageing Society (part of the Studies in Health Technology and Informatics series) Vol. 246, p. 24-28
- Full Text: false
- Reviewed:
- Description: Background: This regional pilot site ‘end-user attitudes’ study explored nurses’ experiences and impressions of using the Nurses’ Behavioural Assistant (NBA) (a knowledge-based, interactive ehealth system) to assist them to better respond to behavioural and psychological symptoms of dementia (BPSD) and will be reported here. Methods: Focus groups were conducted, followed by a four-week pilot site ‘end-user attitudes’ trial of the NBA at a regional aged care residential facility (ACRF). Brief interviews were conducted with consenting nursing staff. Results: Focus group feedback (N = 10) required only minor cosmetic changes to the NBA prototype. Post pilot site end-user interview data (N = 10) indicated that the regional ACRF nurses were positive and enthusiastic about the NBA, however several issues were also identified. Conclusions: Overall the results supported the utility of the NBA to promote a person centred care approach to managing BPSD. Slight modifications may be required to maximise its uptake across all ACRF nursing staff.
A prioritized objective actor-critic method for deep reinforcement learning
- Authors: Nguyen, Ngoc , Nguyen, Thanh , Vamplew, Peter , Dazeley, Richard , Nahavandi, Saeid
- Date: 2021
- Type: Text , Journal article
- Relation: Neural Computing and Applications Vol. 33, no. 16 (2021), p. 10335-10349
- Full Text: false
- Reviewed:
- Description: An increasing number of complex problems have naturally posed significant challenges in decision-making theory and reinforcement learning practices. These problems often involve multiple conflicting reward signals that inherently cause agents’ poor exploration in seeking a specific goal. In extreme cases, the agent gets stuck in a sub-optimal solution and starts behaving harmfully. To overcome such obstacles, we introduce two actor-critic deep reinforcement learning methods, namely Multi-Critic Single Policy (MCSP) and Single Critic Multi-Policy (SCMP), which can adjust agent behaviors to efficiently achieve a designated goal by adopting a weighted-sum scalarization of different objective functions. In particular, MCSP creates a human-centric policy that corresponds to a predefined priority weight of different objectives. Whereas, SCMP is capable of generating a mixed policy based on a set of priority weights, i.e., the generated policy uses the knowledge of different policies (each policy corresponds to a priority weight) to dynamically prioritize objectives in real time. We examine our methods by using the Asynchronous Advantage Actor-Critic (A3C) algorithm to utilize the multithreading mechanism for dynamically balancing training intensity of different policies into a single network. Finally, simulation results show that MCSP and SCMP significantly outperform A3C with respect to the mean of total rewards in two complex problems: Food Collector and Seaquest. © 2021, The Author(s), under exclusive licence to Springer-Verlag London Ltd. part of Springer Nature.
Scalar reward is not enough JAAMAS Track
- Authors: Vamplew, Peter , Smith, Benjamin , Källström, Johan , Ramos, Gabriel , Rădulescu, Roxana , Roijers, Diederik , Hayes, Conor , Heintz, Frederik , Mannion, Patrick , Libin, Pieter , Dazeley, Richard , Foale, Cameron
- Date: 2023
- Type: Text , Conference paper
- Relation: 22nd International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2023, London, 29 May to 2 June 2023, Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS, Vol. 2023-May, p. 839-841
- Full Text: false
- Reviewed:
- Description: Silver et al. [14] posit that scalar reward maximisation is sufficient to underpin all intelligence and provides a suitable basis for artificial general intelligence (AGI). This extended abstract summarises the counter-argument from our JAAMAS paper[19]. © 2023 International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.