11319 research outputs found
Sort by
New Zealand Deprivation Index 2018 - TA68: Waitaki District
No full textFor further information about data sources, interpretation of the graphs, and cautions, please see the separate Introduction Chapter
All data relating to the 2018 census is provided by Stats NZ, https://www.stats.govt.nz/
Ensemble learning through cooperative evolutionary computation
Get PDFBuilding ensembles of classifiers is an active area of research for machine learning, with the fundamental goal of combining the predictions of multiple classifiers to improve prediction accuracy over an individual classifier. In theory, combining classifiers in an ensemble can improve the prediction results by compensating for individual classifier weaknesses in certain areas by benefiting from better accuracy of the other individuals in the same area. Typical ensemble learning approaches require extensive amounts of computation to train and combine multiple models into a single solution. A key question in ensemble learning is: given the total computational effort roughly equivalent to a single monolithic solution, can an ensemble learner achieve comparable or better performance?
In this thesis, a comparison is made between a single complex monolithic agent and an ensemble of many simpler agents that is evolved using equivalent computational effort. To do this, a framework is constructed that enables the comparison of a monolithic approach using complex agents, and an ensemble approach made up of simple agents. This is then applied to the application of buying and selling stocks on a simulated stock market and comparing the results of the two approaches to classify stock data on when to buy and sell. The framework involves creating a population of agents. These are “decision making agents” (DMAs), which evaluate a data source and decide at each time step whether to trade or hold a stock.
In many learning problems, such as the stock trading example used in this thesis, the suitability of a model is measured at a macroscopic level aggregated over multiple decision actions. These problems are not well-suited to traditional learning methods, so evolutionary computation (EC) is frequently used to build machine learning models in these situations. Historically, most EC approaches use a single population to evolve a single solution. A more recent branch of EC research emphasises the use of cooperative coevolution, where the required solution is decomposed into several sub-components, and multiple populations are used in parallel to simultaneously evolve these. There are strong analogies between the divide and conquer strategies of cooperative co-evolution and the building of ensembles in traditional machine learning.
In this thesis, a cooperative co-evolution approach using genetic programming to evolve individual populations and combined them as an ensemble is used to evolve a solution. The agents in the individual populations are evolved with a standard genetic programming approach, where our DMAs are decision trees made up of logic operators (function primitives) and stock indicators (terminal primitives). DMAs are used for both monolithic and ensemble algorithms, but the size of the DMAs varies and the way they are evaluated is different. With the monolithic approach only a single population is used, but the agents in the population evolve to have increased complexity compared to the agents in the ensemble approach. With the cooperative co-evolution ensemble approach, n populations are created and evolved independently, but they are evaluated together using majority voting. The agents used in the ensemble approach are only allowed 1/n of the nodes that the monolithic agent can have, reducing the ensemble’s total complexity to a similar level to that of the monolithic approach.
With this framework, this thesis suggests that an ensemble of simple agents using variance reduction performs as well, and in most cases better than, a complex monolithic agent. The variance reduction process is like that of bagging, with majority voting within the ensemble damping down the behaviour of over-active, risky models to reduce the error component attributable to these risky actions. This variance reduction behaviour was not by design, but was an emergent property. The robustness of these findings is examined under multiple conditions, which include key parameters pertaining to ensemble learning. These include population size and ensemble size, which are examined in this work to gain insights into an optimal set of parameter values.
To ensure that the insights into ensemble learning generalise beyond that of the examined stock trading problem, an alternative unrelated problem, suitable for a cooperative approach is then tested in a similar way. This is the Tartarus problem, in which agents use finite state machines (FSM) for internal states. Previous work in using cooperative co-evolutionary methods on the Tartarus problem focused on decomposition of a single FSM and met with limited success. The co-evolutionary approach used here builds an ensemble of smaller FSMs, each voting on the best action to take. This configuration reduces the computational effort in the mutation operator, therefore allowing an ensemble with more total states to be used for the same overall computational effort. In this context, this approach improves on previous cooperative research and shows that some findings are transferable between applications when using the ensemble approach shown in this research
The role of D2 and CP43 proteins in the bicarbonate-binding environment of Photosystem II
Get PDFIn Photosystem II (PS II), absorbed light energy is used to transfer electrons from water to a primary plastoquinone electron acceptor (QA) and then to the secondary plastoquinone acceptor (QB). A non-heme iron (NHI) is located between QA and QB and ligated by His residues from the D1 and D2 reaction centre protein; in addition, a bicarbonate ion forms a bidentate ligand to the NHI. Stabilisation of bicarbonate is provided by D2-Tyr244, D1-Tyr246 and the two water molecules, W622 and W582, through hydrogen bonds (PDB 4UB6). These two water molecules and amino acid residues around the bicarbonate-binding environment have been hypothesised to participate in stabilisation of a hydrogen-bond network for delivering protons via the bound HCO3- to the QB2-(H+) form of the secondary plastoquinone electron acceptor of PS II. The main purpose of this study was to investigate the importance of D2 and CP43 amino acid residues in maintaining the hydrogen-bond network to provide the stabilisation of bicarbonate and their roles in the structure, electron transfer and protonation steps in the QA-NHI-QB complex of PS II. Mutations targeting four amino acid residues from D2 protein (D2-Glu242, D2-Thr243, D2-Tyr244 and D2-Lys264) and one amino acid residue from CP43 protein (CP43-Arg448) were selected. Nearly all mutants showed impaired oxygen evolution and were highly susceptible to photodamage in the presence of PS II-specific electron acceptors; however, they were able to acclimate to high light when HCO3– was added. In addition, acceptor side electron transfer was altered in these strains, with a decrease in the forward electron transfer between QA and QB, as well as the back reaction from QA, in the presence of the PS II-specific inhibitor 3,4-dichloro-1,1-dimethyl urea (DCMU). Furthermore, all mutants except the D2-Glu242 mutant exhibited the accumulation of unassembled complexes containing the CP43 subunit and an ineffective repair mechanism was also observed in these mutants that showed delayed protein synthesis following the high-light treatment based on 35S-Met protein labelling experiments. This study indicated that the disruption of the hydrogen-bond network associated with the bicarbonate-binding environment impairs PS II assembly, electron transfer and also the repair mechanism of the photosystem following high-light-induced damage
The spatio-temporal dynamics of fish and invertebrate communities within coastal wetland systems
Get PDFThis thesis investigates the spatial and temporal dynamics of fish and invertebrate populations within coastal wetlands in Southland, New Zealand. Each chapter focusses on a different aspect of the community and its interaction with the environment. Initially fish in four systems were surveyed over two years, where great variation in abundance and diversity of species was observed, particularly in relation to connectivity to other waterways (Chapter 2). During the survey period, significant dewatering events occurred; the duration of these low water events had a significant impact on how rapidly the community recovered. Following this was an in-depth investigation of the invertebrate community dynamics over a spring to autumn sampling period, focussed in one of the initial wetland areas, a constructed wetland system with many small interconnected ponds (Chapter 3). A progression of dominant microcrustaceans were observed as was a succession of larger more predatory taxa. Spatial variation was also observed, related to hydrological connectivity and nutrient levels. An investigation into īnanga population dynamics was also conducted, as they are a key species in this system and there is a notable gap in knowledge on their habits within small ponds (Chapter 4). This found spring-hatching fish with a faster growth rate than those that hatch later in the year. Diet preferences indicate both benthic and pelagic foraging, and the presence and consumption of microplastic filaments. The study culminates in an in-depth investigation of the recovery of both fish and invertebrate communities after a drought within one system, tying together observations from the other chapters (Chapter 5). This highlighted the importance of spatial heterogeneity and connectivity for rapid invertebrate recovery, as well as the value of deep channels that are more resistant to drying in supporting fish recolonization. Overall, this research has provided insights into communities of fish and invertebrates, generating new information relevant to the specific systems and species as well as the wider body of knowledge on wetlands
Local government and youth voter turnout: Obstacles and solutions for Aotearoa New Zealand
Get PDFLocal government youth voter turnout in New Zealand is very low. Although it is not too dissimilar to other western democracies where voting is not compulsory, the low rates are ineffective in terms of a healthy democracy. Low voter turnout amongst 18- to 24-year-olds can undermine the political process. Research has shown that youth are not turning out to vote in local government elections, mainly due to a lack of available information, a feeling that local government is a second-order election, and because they feel they have no influence in the political process. This thesis asks “Do young people (18-24 years think that eVoting, compulsory voting, and a reduction in the voting age would encourage them to vote in New Zealand local government elections?”
This research explores the causes of low youth voter turnout for 18- to 24-year-olds in local government elections in Dunedin and Palmerston North, New Zealand, as described by young voters themselves. The research explores, through a survey and focus groups, the motivations of young people who did turn out to vote. The focus here is on young voters in two specific cases—Dunedin and Palmerston North—both with high percentages of potential youth voters at tertiary institutions. The thesis proposes and considers the effectiveness (in theory and practice) of three solutions to boost youth turnout: 1) introducing eVoting; 2) lowering the voting age to 16 years; and 3) moving to compulsory voting. Insights into youth not feeling that they had sufficient information to make informed decisions when voting were unearthed and explored.
The research draws the following conclusions: 1) eVoting appeals to young voters and could potentially increase youth voter turnout; 2) young people are divided about introducing compulsory voting, particularly at local government level; 3) youth are opposed to reducing the voting age to 16 because they feel that it is too young. Whilst not initially set as a focus of the current research, civics education was raised by so many participants in both the surveys and the focus groups, that it needs to be addressed. The current research limitations, including the need to explore the generalisability of the results across New Zealand and ideas for future research directions, are also covered
Ontogenetic differences in behaviour of the Otago skink (Oligosoma otagense) at Ōrokonui Ecosanctuary
Get PDFPersonalities of animals may demonstrate ontogenetic changes in response to a plethora of different environmental factors and experiences early in life. Understanding these ontogenetic personality changes aids in understanding how individuals tolerate, act and react to environments, conspecifics and other species throughout their lifespan. This behavioural topic remains relatively unexplored for many reptilian species. Because non-avian reptiles are such a diverse group of vertebrates in, ecology, behaviour and morphology, more research on ontogenetic personality changes is required for this class of animals.
Relatively little is known about the social behaviour and personalities of the Otago skink (Oligosoma otagense), or how they differ between age-groups. This study aimed to examine (1) whether Otago skinks demonstrated repeatability of behaviours over four sample periods, and therefore have personalities, (2) whether sociality, aggression, boldness and exploration levels differed between adult, sub-adult and juvenile Otago skinks, and (3) whether temporal variation of sociality, aggression, boldness and exploration variables differed between adult, sub-adult and juvenile Otago skinks over four sample periods. This study was conducted on the translocated Otago skink population at Ōrokonui Ecosanctuary from the 15th of January until the 6th of October, 2020. Photographic surveys and behavioural assays were used to examine behaviours that represented boldness, sociality, aggression, and exploration levels of each skink.
Individual repeatability was low for all measured behaviour variables (R < 0.120), with only boldness (proportion of sessions sighted in), exploration (proportion of sightings present at a favourite tor) and sociality (proportion of sightings present in a group) variables having significant repeatability estimates. Individual repeatability of the behavioural assay boldness variables were even lower (R < 0.104), and none of these low repeatabilities were significant. This therefore suggests that no personalities’ were detected for this Otago skink population, however this does not mean that they were not there.
Several behaviours showed significant differences between age-groups. Adults were sighted significantly more often than juveniles in photographic surveys, suggesting that adults were bolder than juveniles. Juveniles were sighted significantly more often at the tor
ii
they spent the most time at and visited fewer tors compared to sub-adults and adults, suggesting that adults and sub-adults were more explorative than juveniles. Juveniles were sighted in larger social groups significantly more often than sub-adults, therefore indicating that juveniles had higher levels of sociality. Aggressive interactions were only observed in adults and sub-adults (not juveniles). This could suggest that sub-adults and adults had higher aggression levels.
Overall there was little difference between the age-groups for correlation of behaviours between sample periods. Age-group had a significant effect on the temporal variation of proportion of sightings present at a favourite tor between two sample periods, where adults demonstrated increased consistency compared to sub-adults. There was also some stability in boldness and exploration behaviours over sample periods for all age-groups.
Overall this study has provided new insights into how sociality, aggression, boldness and exploration levels differ between the different age-groups of Otago skinks at Ōrokonui Ecosanctuary. Juveniles were more social and less aggressive, bold and explorative than adults and sub-adults. This study has also revealed that this Otago skink population have relatively low repeatability and stability of behaviours. These findings will provide further insight into how Otago skink individuals will differ in how they act and react to different environments, situations and conspecifics at different points in their lifespan. This information may also aid in the future translocations and set-up of enclosures for the existing natural populations of this endangered species
A shaky foundation? John Owen and Thomas F. Torrance on Christ's mediation
Get PDFThe goal of this dissertation is to explore how the theology proper of both John Owen and Thomas F. Torrance informs their respective accounts of Christ’s mediation. To this end, the primary question behind the entire investigation is how Owen and Torrance conceptualize God and all things in relation to God. I have chosen Owen and Torrance as dialogue partners because they offer us two different approaches to ‘God and all things in relation to God’ from within the same broadly Reformed tradition. Furthermore, both are also catholic theologians, engaging with, and appropriating sources throughout the ancient, medieval, Reformation, and modern Christian traditions, and in the case of Torrance, the traditions of both East and West. In addition, both give significant attention to Christ’s priesthood. This emphasis on Christ, combined with their preoccupation with God, allows us to see how both Owen and Torrance move from God in himself to God for us. Furthermore, Owen’s scholastic and broadly Thomistic theology is what Torrance seeks to avoid at every turn. Therefore, by placing Owen and Torrance in dialogue, I test the ‘Calvin versus the Calvinists’ thesis that is by and large disproven today by historians showing both the continuity and discontinuity of Calvin with his heirs. The Reformed shift to a soteriological reading of theology proper – as I will develop in this study – means, I argue, that the relationship of one Reformed generation to another is more complex than both sides suggest.
After investigating both thinkers’ understanding of theology proper, God as one and God as tripersonal, I then show how the doctrine of God informs Christ’s hypostatic union and Christ as mediator. The main body of the work consists of understanding each thinker on his own terms, though critiques, mainly related to conflations of God in se and God for us, are offered along the way. Throughout the study, I argue that in both Owen and Torrance various conflations arise that are all, in some way, related to their soteriological reading of God in se. By conflation, I simply mean that something outside of God in himself – anything other than God’s eternally perfect being – becomes the foundation for our understanding of God. I show that in both Owen and Torrance, soteriology, at least at points, becomes the foundation for theology proper as God’s saving works in Christ become determinative for our understanding of God. All of this builds toward a constructive conclusion that calls for a retrieval of God’s goodness as the solution to avoiding conflation in our theological systems. While less intentional and radical in Owen, and more purposely revisionary in Torrance, the turn toward a soteriological foundation is one that places theology proper on an insecure foundation.
In chapter 1, I investigate Owen’s understanding of God’s being and attributes. In chapter 2, I turn to Owen’s Trinitarian theology with particular attention to the divine processions and missions showing that Owen’s use of the covenant of redemption leads him to inadequately distinguish God ad intra and God ad extra. In chapter 3, I show that Owen’s understanding of the hypostatic union and Christ’s two natures follows Chalcedon and Aquinas and a Thomistic understanding of God in se and God for us, though his understanding of the role of the Holy Spirit upon Christ’s human nature reveals his continual struggle to relate God in se and God for us. Finally, in chapter 4, I show that Owen’s theology of Christ as mediator seeks to hold together Aquinas’s metaphysically motivated Christology and Calvin’s office Christology. In chapter 5, I show how Torrance’s understanding of God’s being is determined by the Son’s incarnation. The economic is the immanent because Jesus Christ is homoousios. In chapter 6, I investigate Torrance’s theology of the three divine persons, arguing that Torrance’s understanding of hypostasis and being pushes his Trinitarian theology in a somewhat modalist direction. In chapter 7, I show that Torrance understands the Son’s mission, not as a communication of God’s goodness, but as a self-communication – that is revelation – that demands a soteriological starting point. Lastly, in chapter 8, I turn to Torrance’s theology of election and Christ’s mediation as the vicarious human being. I argue that Torrance’s construction aligns Christ’s divinity too closely with Christ’s humanity. By concluding with a retrieval of God’s goodness, I advance our understanding of how Christ is predestined for his redemptive mission without letting soteriology slip into our foundation, theology proper
Autumn Meal Plans
No full textEight sets of meal plans, which are designed to feed a flat of four students five dinner meals per week. The meals are healthy, affordable, and zero-waste. As well as recipes and shopping lists, the meal plans contain helpful information such as nutrition information, the cost per serving and food waste reduction tips
Discounts on Reparations for Lost Earnings
Get PDFAnalyses the case law regarding discounts for lost earnings. Addresses three forms of discount that courts have applied in some cases: discounts for up-front payment; because other ACC entitlements have been paid; and for the benefit of avoiding litigation.Peer Reviewe
Agony, Exclusion and Colonial Reproduction: A Critical Examination of the Doctrine of Difference in Aotearoa New Zealand
Get PDFThis article examines how contemporary legal discourse perpetuates and reproduces colonial structures and some less risky alternatives. It does so through an inquiry into how First Peoples enter into legal relationships with the Crown. In this inquiry, we combine the doctrine of discovery, a well-known conception the literature of Indigenous peoples and law, with the “dynamic of difference”, a concept stemming from ‘Third World Approaches to International Law’, to generate what we call the “doctrine of difference”. The doctrine of difference aids in orienting legal discourse within broader conceptualisations of power to better understand the risksthat arise when First Peoples enter into legal relationships with the Crown. One way of entering into legal relationships is to use rights-based legal discourse. Through the doctrine of difference, we argue that this requires rights claimants to construct themselves as subjects of a pre-existing legal discourse, which reproduces the universalising and naturalising dynamics associated with colonialism. In contrast, First Peoples may enter into a legal relationship through an agonistic deliberative process that re-centres and invests in place-based legalities, languages and forms of life. Such agonistic deliberative alternatives may partially avoid some of the universalising and naturalising effects (re)produced by rights-based legal discourse.Peer Reviewe