The MSc programme aims to teach conservation as a dynamic discipline integral to all the major areas of human concern - judicial reform, political economy, spatial planning, poverty alleviation, human and institutional capacity, agriculture, and population growth, in addition to the hard science of biodiversity and ecology. This grant supports 4 full studentships for three years.

In recent years, there has been a lot of research focused on a new emerging field called Meta-Learning. The idea of Meta-Learning can be traced back to Donald B. Maudsley's work in 1979, who referred to it as "the process by which learners become aware of and increasingly in control of habits of perception, inquiry, learning, and growth that they have internalized". Hence the goal of Meta-Learning is to understand how one can grasp ideas from data and transfer these concepts to similar tasks efficiently. However, a lot of machine learning(ML) models these days require a lot of data and training time to get good performances, i.e. Deep learning models. Meta-Learning aims to find the underlying structure of these tasks. So given a new task the model is able to adapt quickly. Understanding this would be a step towards Artificial General Intelligence. There has been a huge interest in Meta-Learning in the recent years, as it allows for efficient learning from multiple dataset and tasks in a joint manner. Among the most popular algorithms, we have "Model Agnostic Meta Learning" (MAML), which is a optimization based model. Another popular model is the Neural Process, which has received a lot of attention recently as it is able to model distributions over functions instead of modelling them over parameters. In this project, we aim to use the Meta-Learning framework for modelling distributions over the expectation of conditional distributions. Density estimation is an important problem in a lot of Machine Learning task, as it allows us, for example, to detect outliers or get a better understanding of the underlying data distribution itself. Being able to compute the log-likelihood of new samples, enables us to discard/further investigate those specific data points, which can be very useful in a wide variety of scenarios such as medicine for example. In addition, uni-model distributions for prediction tasks are often used as they allow for simple computations such as the popular Kullback-Leibler divergence which measures the distance between distributions. However, unimodality is a strong assumption that is usually not the case in real-world scenarios. Hence, we propose to firstly model the expectation of the conditional distribution without the unimodality constraint in a non parametric fashion and secondly, learn representations of distributions that can be transferred across different density estimation tasks. The novel methodology, we want to apply to this task can be quantified under the umbrella of Kernel methods. We make use of the well-established methodology of Kernel mean embeddings, which allows us to encode a distribution in a functional space. This space can then be learned to be optimal for transferring structures between datasets. A novel way of tackling this learning problem is to use "Constrastive Noise Estimation (NCE)", which is a recently developed method that allows us to learn representations that encode the underlying structure in a signal, thereby ignoring low-level information such as noise. In essence, the idea consists of comparing your data with noise and thus learning a suitable function space that is able to distinguish them. This method has shown impressive results in a lot of different areas such as speech, images, text and reinforcement learning. Due to its computational scalability and representative power, we aim to use NCE for learning the function space of conditional distribution and thus learn a useful representations for density estimation across different datasets. Hence, the primary aim of this project is representation learning of conditional distributions, that captures the underlying structure of the densities, using the recently developed and promising technique of NCE. This project falls within the EPSRC Information and communication technologies (ICT) and Engineering research area. I currently do not have any companies or collaboarators involved in my projects.

I study two areas of the mammalian brain known as amygdala and hippocampus. These brain regions are of fundamental importance in cognitive functions such as learning, memory, fear and emotional states in health and psychiatric diseases. The amygdala and the hippocampus are composed of several distinct types of nerve cells. About 20% of them release and use a substance called GABA to communicate with other nerve cells. This neurotransmitter normally reduces the activity of nerve cells and plays key roles in the normal brain and in the diseased brain. To define the functions of the amygdala and hippocampus under normal conditions and in experimental models of psychiatric diseases, it is important to characterise the types of GABA cells, their structure, and their molecular properties. My work identifies specific types of GABA cells which have clear roles in behaviour and are related to psychiatric disorders and the study of them can lead to the identification exciting novel principles of how neurons communicate with each other. For example, I study a type of nerve cell in the amygdala which is of paramount importance for fear extinction, a process that extinguishes fear once fear associations have been learned. Fear extinction is the corner-stone of the psychological therapy of anxiety because the inability to extinguish fear responses is a trait of several anxiety disorders. As the amygdala and hippocampus are associated with post-traumatic stress disorders and other anxiety conditions, this project may also advance not only the knowledge of fundamental neuroscience, but also help to elucidate the mechanisms of action of widely used drugs such as anaesthetics, anticonvulsants and anxiolytics.

The most severe form of human malaria is caused by the Plasmodium falciparum parasite. Despite recent and encouraging advances in malaria control measures, current estimates suggest that in 2015 there were still over 200 million clinical cases leading to 438,000 deaths. Consequently, the development of an effective and durable vaccine remains a key strategic goal to aid the control, local elimination and eventual eradication of this disease. The mainstay approach to vaccination against the blood-stage of malaria infection is to induce antibodies against the merozoite form of the parasite that invades red blood cells (RBC). Such a vaccine would protect against disease severity and could reduce transmission. A major recent advance has been the identification of a critical non-redundant interaction during RBC invasion - mediated between basigin (CD147) on the RBC surface and the parasite protein RH5. New data suggest RH5 is delivered to the parasite's surface in a protein complex whose components are not fully elucidated. All known components of the complex are essential and need to be investigated as vaccine targets to complement on-going approaches against RH5. RH5 Interacting Protein (RIPR) is one of the proteins that form this protein complex. Initial studies have suggested that RIPR may make an effective vaccine. However, RIPR is a large and difficult protein to manufacture. The primary aim of this project is to generate the pre-clinical clinical data which supports the development of a RIPR vaccine. This includes identifying neutralising epitopes on the RIPR protein, investigating novel methods of manufacturing RIPR and testing new vaccine technologies such as Viral Like Particles (VLPs) to boost the immune response. In parallel, a second project will be to develop monoclonal antibodies against RIPR. There are very few published anti-RIPR monoclonal antibodies (mAbs) and our group has so far been unsuccessful at generating neutralising anti-RIPR mAbs. Generating new neutralising mAbs will aid structure-based vaccine design by providing powerful tools for studying neutralising epitopes and the interactions between proteins in the complex. In addition, mAbs may be engineered to be powerful therapeutics in their own right, such therapeutics may be important for future malaria treatment strategies. This DPhil project will have two overriding aims: 1. To thoroughly investigate RIPR as a candidate malaria vaccine. 2. To develop monoclonal antibodies (mAbs) against RIPR. These will be screened for functional anti-parasitic activity and combined with leading mAbs against RH5 to identify a highly inhibitory mAb cocktail.