David Powers | Flinders University of South Australia (original) (raw)
Books by David Powers
A method for generating an index of data available from a server, including processing data on t... more A method for generating an index of data available from a server, including processing data on the server to access data items for a central index, the data items including network addresses and terms, compiling an index file including the data items, and transmitting the index file to the central index. The processing may include locating database query statements in the data, and the data items then include input tuples for the statements. The index is accessible from
servers, and includes page entries including a program address for a program for generating a dynamic page and
input tuples for submission to the program to generate the page, and search entries identifying the dynamic pages and
identifying the tuples corresponding to search terms. A search engine operable on the index, is able to access the
Search entries to identify dynamic pages Corresponding to search terms of a search query, and access the page entries to generate addresses for the dynamic pages identified, the addresses being generated on the basis of the program address and the tuples.
Our subject is the acquisition of Natural Language (NL) by computers. NL is not, in our view, a s... more Our subject is the acquisition of Natural Language (NL) by computers. NL is not, in our view, a surface expression, or epiphenomenon, of a deeper, underlying cognitive process in the human brain. It is rather fundamental to, and pervasive of, cognition itself. For this reason we think that language is not the sole preserve of linguistics, but is pivotal in all our interactions with the world, in our science, and in our thought.
Papers by David Powers
Soft Computing, Jan 7, 2020
arXiv (Cornell University), Mar 22, 2015
Journal of Intelligent & Robotic Systems, 2018
2015 IEEE International Symposium on Robotics and Intelligent Sensors (IRIS), 2015
Journal of Marine Science and Application, 2018
International Conference on Computers in Education, 2017
13th Australasian International Conference on Speech Science and Technology, 2010
Under an ARC Linkage Infrastructure, Equipment and Facilities (LIEF) grant, speech science and te... more Under an ARC Linkage Infrastructure, Equipment and Facilities (LIEF) grant, speech science and technology experts from across Australia have joined forces to organise the recording of audio-visual (AV) speech data from representative speakers of Australian English in all capital cities and some regional centres. The Big Australian Speech Corpus (the Big ASC) will provide a standard recording setup and a collaboratively-designed elicitation protocol to create a corpus of AV speech data incorporating annotations and ...
Visual relationship detection aims to understand real-world interactions between object pairs by ... more Visual relationship detection aims to understand real-world interactions between object pairs by detecting visual relation triples written in the form of (subject, predicate, object). Previous work has explored the use of contrastive learning to generate joint visual and language embeddings that aid the detection of both seen and unseen visual relation triples. However, these contrastive approaches often learned the mapping functions implicitly and did not fully consider the underlying structure of visual relation triples, limiting the models’ use cases and their ability to generalize to unseen compositions. This ongoing work aims to construct joint visual and language embedding models that can capture such hierarchical structure between objects and predicates by explicitly imposing structural loss constraints. In this short paper, we propose VLTransE, a novel embedding model that applies translational loss in conjunction with the visual-language contrastive loss to learn transferab...
ArXiv, 2016
This paper presents a hybrid route-path planning model for an Autonomous Underwater Vehicle's... more This paper presents a hybrid route-path planning model for an Autonomous Underwater Vehicle's task assignment and management while the AUV is operating through the variable littoral waters. Several prioritized tasks distributed in a large scale terrain is defined first; then, considering the limitations over the mission time, vehicle's battery, uncertainty and variability of the underlying operating field, appropriate mission timing and energy management is undertaken. The proposed objective is fulfilled by incorporating a route-planner that is in charge of prioritizing the list of available tasks according to available battery and a path-planer that acts in a smaller scale to provide vehicle's safe deployment against environmental sudden changes. The synchronous process of the task assign-route and path planning is simulated using a specific composition of Differential Evolution and Firefly Optimization (DEFO) Algorithms. The simulation results indicate that the propose...
In this paper we introduce a novel method of automating thesauri using syntactically constrained ... more In this paper we introduce a novel method of automating thesauri using syntactically constrained distributional similarity. With respect to syntactically conditioned cooccurrences, most popular approaches to automatic thesaurus construction simply ignore the salience of grammatical relations and effectively merge them into one united 'context'. We distinguish semantic differences of each syntactic dependency and propose to generate thesauri through word overlapping across major types of grammatical relations. The encouraging results show that our proposal can build automatic thesauri with significantly higher precision than the traditional methods.
Proceedings 24th Australian Computer Science Conference. ACSC 2001
2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), Nov 1, 2016
A method for generating an index of data available from a server, including processing data on t... more A method for generating an index of data available from a server, including processing data on the server to access data items for a central index, the data items including network addresses and terms, compiling an index file including the data items, and transmitting the index file to the central index. The processing may include locating database query statements in the data, and the data items then include input tuples for the statements. The index is accessible from
servers, and includes page entries including a program address for a program for generating a dynamic page and
input tuples for submission to the program to generate the page, and search entries identifying the dynamic pages and
identifying the tuples corresponding to search terms. A search engine operable on the index, is able to access the
Search entries to identify dynamic pages Corresponding to search terms of a search query, and access the page entries to generate addresses for the dynamic pages identified, the addresses being generated on the basis of the program address and the tuples.
Our subject is the acquisition of Natural Language (NL) by computers. NL is not, in our view, a s... more Our subject is the acquisition of Natural Language (NL) by computers. NL is not, in our view, a surface expression, or epiphenomenon, of a deeper, underlying cognitive process in the human brain. It is rather fundamental to, and pervasive of, cognition itself. For this reason we think that language is not the sole preserve of linguistics, but is pivotal in all our interactions with the world, in our science, and in our thought.
Soft Computing, Jan 7, 2020
arXiv (Cornell University), Mar 22, 2015
Journal of Intelligent & Robotic Systems, 2018
2015 IEEE International Symposium on Robotics and Intelligent Sensors (IRIS), 2015
Journal of Marine Science and Application, 2018
International Conference on Computers in Education, 2017
13th Australasian International Conference on Speech Science and Technology, 2010
Under an ARC Linkage Infrastructure, Equipment and Facilities (LIEF) grant, speech science and te... more Under an ARC Linkage Infrastructure, Equipment and Facilities (LIEF) grant, speech science and technology experts from across Australia have joined forces to organise the recording of audio-visual (AV) speech data from representative speakers of Australian English in all capital cities and some regional centres. The Big Australian Speech Corpus (the Big ASC) will provide a standard recording setup and a collaboratively-designed elicitation protocol to create a corpus of AV speech data incorporating annotations and ...
Visual relationship detection aims to understand real-world interactions between object pairs by ... more Visual relationship detection aims to understand real-world interactions between object pairs by detecting visual relation triples written in the form of (subject, predicate, object). Previous work has explored the use of contrastive learning to generate joint visual and language embeddings that aid the detection of both seen and unseen visual relation triples. However, these contrastive approaches often learned the mapping functions implicitly and did not fully consider the underlying structure of visual relation triples, limiting the models’ use cases and their ability to generalize to unseen compositions. This ongoing work aims to construct joint visual and language embedding models that can capture such hierarchical structure between objects and predicates by explicitly imposing structural loss constraints. In this short paper, we propose VLTransE, a novel embedding model that applies translational loss in conjunction with the visual-language contrastive loss to learn transferab...
ArXiv, 2016
This paper presents a hybrid route-path planning model for an Autonomous Underwater Vehicle's... more This paper presents a hybrid route-path planning model for an Autonomous Underwater Vehicle's task assignment and management while the AUV is operating through the variable littoral waters. Several prioritized tasks distributed in a large scale terrain is defined first; then, considering the limitations over the mission time, vehicle's battery, uncertainty and variability of the underlying operating field, appropriate mission timing and energy management is undertaken. The proposed objective is fulfilled by incorporating a route-planner that is in charge of prioritizing the list of available tasks according to available battery and a path-planer that acts in a smaller scale to provide vehicle's safe deployment against environmental sudden changes. The synchronous process of the task assign-route and path planning is simulated using a specific composition of Differential Evolution and Firefly Optimization (DEFO) Algorithms. The simulation results indicate that the propose...
In this paper we introduce a novel method of automating thesauri using syntactically constrained ... more In this paper we introduce a novel method of automating thesauri using syntactically constrained distributional similarity. With respect to syntactically conditioned cooccurrences, most popular approaches to automatic thesaurus construction simply ignore the salience of grammatical relations and effectively merge them into one united 'context'. We distinguish semantic differences of each syntactic dependency and propose to generate thesauri through word overlapping across major types of grammatical relations. The encouraging results show that our proposal can build automatic thesauri with significantly higher precision than the traditional methods.
Proceedings 24th Australian Computer Science Conference. ACSC 2001
2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), Nov 1, 2016
Applied Soft Computing, 2018
Australian New Zealand Conference on Intelligent Information Systems, 1996
This paper deals with the problem of processing acoustic signals originating from multiple source... more This paper deals with the problem of processing acoustic signals originating from multiple sources in a potentially noisy environment. Previous research in speech processing and cognitive modelling has tended to concentrate on single sources and relatively noise-free signals. Separating out dif ferent signals from a multitude of sources is a significant part of human auditory processing. In speech processing research,
Hanieh Bakhshayesh, B.Eng. Honours Thesis, Flinders University, 2012
In this project, well-known synchronization measures (coherence, correntropy, cross correlation, ... more In this project, well-known synchronization measures (coherence, correntropy, cross correlation, phase synchrony and GFS) were compared to understand which one is the best measure for applying to EEG signals (electrical activity recorded from the human scalp).
EEG is a nonstationary and noisy signal therefore a good synchronization measure should be insensitive to noise, be able to detect both linear and nonlinear relationships, and be able to detect nonstationary relationships. Time resolution and frequency resolution are other important properties of synchronization measures.
Different approaches were considered to achieve the above requirements. Synchronization measures were applied to two types of simulated data: noisy sinusoidal signals with linear relationships and unidirectional coupled Hénon maps with nonlinear relationships.
Different measures have both good and bad features. No measure is the clear winner but correntropy and phase synchrony shows promising results. The most appropriate synchronization measure must be chosen with knowledge of the application in hand.
Additionally, the synchronization measures were applied to real EEG data. The first application was to find a synchronization measure which may be able to identify a persistent effect of disease in the brain. It is hypothesised that there are changes in the synchronization of brain signals of a patient with certain diseases, which are persistent across tasks. Each synchronization measure was evaluated for different subjects across different tasks. Phase synchrony and then coherence have the smallest dispersion from the mean (lowest standard deviation) and can be considered as the measures which are the least sensitive to changes due to the task. As phase synchrony also performed well on simulated data, it may be an appropriate measure to detect a persistent effect of disease in the brain.
Thinking was the second application considered with real EEG. The main feature of thinking is its short time scale; therefore it is required to find a measure which is able to detect rapid changes in brain activities. To achieve this, all synchronization measures were applied to short blocks of EEG, sliding the block one sample at a time. There was no evidence that this analysis is able to distinguish EEG from noise. In another experiment, synchronization measures were applied to blocks of data of decreasing size, to identify measures that can detect synchronisation over very short timescales (i.e. thoughts) where synchrony is not apparent at long time scales. None of the results was in agreement with the expected outcome.
As phase synchrony performed well on simulated data and robustly meets the null hypothesis of real EEG, in overall it can be concluded that it is an appropriate synchronization measure for further experiment on brain’s signal .but still it should be noted, application and aim of research are important aspects that needs to be consider for choosing the preferable synchronization measure.
Corey Taylor, LitReview, M.InfoTech, Flinders University