Theory and methodologies for solving the common sense of intelligence

Lead Research Organisation: University of Oxford
Department Name: Statistics

Abstract

This project falls within the EPSRC Mathematical sciences research area.
This project lies in the context of computational statistics and in particular statistical machine learning, deep learning and Bayesian nonparametrics.
The number of data generated every day is steadily increasing. Analyzing and understanding this large volume of information is important. Indeed, creating algorithms so that computers can learn from the data they are given without any human intervention can give faster and more accurate results. This can have a lot of applications in our society and impact many areas socially and economically such as healthcare, transports, security ... For instance, in medicine, intelligent systems can assist doctors in their diagnosis through the analysis of radiographs. Another example is the analysis of images that can help policemen arrest individuals wanted. Consequently, developing both theoretical and computational methods and training systems to process the quantity and diversity of data we have is thus a topical and important challenge to be solved. So far, some systems have been successfully experimented to perform some very specific tasks. However, the common sense of intelligence, that is the perception and understanding of the world in which we are living in and that a child generally acquires by the age of two, has not been solved yet. The project would like to explore this question.
The aim of the project is to contribute to both the modelling part of statistical machine learning and the development of learning algorithms, that are essential for Artificial Intelligence in general.
The project plans to research the intersection between deep learning and bayesian methods. Deep learning and kernel methods have proven to be very flexible methods and should therefore be interesting in developing methodologies based on heterogeneous or large scale data. Bayesian nonparametrics have proved effective in modelling some real-world situations such as networks.
Studying the relevance of probabilistic programming in Bayesian nonparametrics is also an objective of the project. Indeed, some libraries for probabilistic inference such as Edward or some probabilistic programming language such as Pyro, recently released by Uber, have recently been developed and incorporated to self-driving car systems, which shows that there is a growing need for probabilistic programming in AI systems.

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
EP/N509711/1 01/10/2016 30/09/2021
2207522 Studentship EP/N509711/1 01/10/2018 30/09/2022 Charline Le Lan