A5025865844- Evolutionary Game Theory and Cooperation
- Animal Behavior and Reproduction
- Evolution and Genetic Dynamics
- Plant and animal studies
- Genetic diversity and population structure
- Topic Modeling
- Experimental Behavioral Economics Studies
- Natural Language Processing Techniques
- Mobile Health and mHealth Applications
- Gestational Diabetes Research and Management
- Bariatric Surgery and Outcomes
- Plant Virus Research Studies
- Forest Insect Ecology and Management
- Speech Recognition and Synthesis
- Animal Ecology and Behavior Studies
- Game Theory and Applications
- Gaussian Processes and Bayesian Inference
- Bayesian Methods and Mixture Models
- Botany and Plant Ecology Studies
- Viral Infectious Diseases and Gene Expression in Insects
- Hippo pathway signaling and YAP/TAZ
- Forest Ecology and Biodiversity Studies
- Insect Resistance and Genetics
- Wnt/β-catenin signaling in development and cancer
- Mathematical and Theoretical Epidemiology and Ecology Models
UNSW Sydney
2012-2017
University of Southern California
2015
Université du Québec à Montréal
2009-2012
Astex Pharmaceuticals
2012
University of Alberta
2007
Usually bilingual word vectors are trained "online". Mikolov et al. showed they can also be found "offline", whereby two pre-trained embeddings aligned with a linear transformation, using dictionaries compiled from expert knowledge. In this work, we prove that the transformation between spaces should orthogonal. This obtained singular value decomposition. We introduce novel "inverted softmax" for identifying translation pairs, which improve precision @1 of Mikolov's original mapping 34% to...
Behavioral ecologists have long been comfortable assuming that genetic architecture does not constrain which phenotypescan evolve (the "phenotypic gambit"). For flexible behavioral traits, however, solutions to adaptive problems are reached only by evolution but also changes within an individual's lifetime, via psychological mechanisms such as learning. Standard optimality approaches ignore these mechanisms, implicitly they do the expression of behavior. This assumption, we dub gambit, is...
Summary Genetic algorithms are a heuristic global optimisation technique mimicking the action of natural selection to solve hard problems, which has enjoyed growing usage in evolution and ecology. However, there is little standardisation how genetic implemented these fields even less practical advice for those hoping implement their own models. This article aims demystify provide assistance researchers; basic programming knowledge important working with algorithms, but none required read...
Associative learning is essential for resource acquisition, predator avoidance and reproduction in a wide diversity of species, therefore key target evolutionary comparative cognition research. Automated operant devices can greatly enhance the study associative yet their use has been mainly restricted to laboratory conditions. We developed portable, weatherproof, battery-operated device conducted first fully automated colour-associative experiment using free-ranging individuals wild. used...
Many viruses, particularly RNA mutate at a very high rate per genome replication. One possible explanation is that mutation rates are selected to meet the challenge of fluctuating environments, including host immune response. Alternatively, recent studies argue viruses evolve under trade-off between replication speed and fidelity such fast selected, and, along with it, rates. Here, in addition these factors, we consider role viral life-history properties: namely, within-host dynamics...
Successful management of gestational diabetes mellitus (GDM) reduces the risk morbidity in women and newborns. A woman's blood glucose readings factors are used by clinical staff to make decisions regarding initiation pharmacological treatment with GDM. Mobile health (mHealth) solutions allow real-time follow-up GDM timely management. Machine learning offers opportunity quickly analyze large quantities data automatically flag at requiring treatment.
Prior to seeking professional medical care it is increasingly common for patients use online resources such as automated symptom checkers. Many systems attempt provide a differential diagnosis based on the symptoms elucidated from user, which may lead anxiety if life or limb-threatening conditions are part of list, phenomenon termed 'cyberchondria' [1]. Systems that advice where seek help, rather than diagnosis, equally popular, and in our view most useful information. In this technical...
Individuals foraging in groups can use two different tactics for obtaining food resources. either search sources themselves (producing) or they join discoveries of others (scrounging). In this study we a genetic algorithm spatially explicit producer-scrounger game to explore how individuals compromise between exploration (an important axis animal personality) and scrounging characteristics the environment affect compromise. Agents varied searched optimal combination scrounging. The...
Niche construction has received increasing attention in recent years as a vital force evolution and examples of niche have been identified wide variety taxa, but viruses are conspicuously absent. In this study we explore how can lead to engineering their hosts (including behavioural manipulation) with feedback on selective pressures for viral transmission virulence. To illustrate concept focus Baculoviridae, family invertebrate that evolved modify the feeding behaviour lepidopteran liquefy...
When animals forage socially, individuals can obtain prey from their own searching (producer tactic) or by using the behaviour of others (scrounger when it provides inadvertent social information (ISI) that food has been located. This ISI may either indicate location (social information, SI), quality resource (public PI). To date, few studies have explored selective consequences for being exploited predators use ISI. Prey such should evolve traits favour high levels (scrounging) because this...
A major goal of molecular evolution is to determine how natural selection has shaped the a gene. One approach taken by methods such as K /K S and McDonald-Kreitman (MK) test compare frequency non-synonymous synonymous changes. These methods, however, rely on assumption that change in one mutation will not affect changes other mutations. We demonstrate linkage between sites can bias measures based Using forward simulation Wright-Fisher process, we show hitch-hiking deleterious mutations with...
Objective: The use of routinely-acquired medical data for research purposes requires the protection patient confidentiality via anonymisation. objective this work is to calculate risk re-identification arising from a malicious attack an anonymised dataset, as described below. Methods: We first present analytical means estimating probability single in k-anonymised dataset Electronic Health Record (EHR) data. Second, we generalize solution obtain multiple patients being re-identified. provide...