Academic Research Insights
Neurotech - windows to your soul
An introduction to Neurotech by Aldo Faisal, Professor of AI & Neuroscience
Dr Aldo Faisal is a neuroscientist seeking to decode how our brains generate behaviour from our intentions. His research has led to the creation of many neurotechnologies, including brain-machine-interfaces that utilise eye tracking, which could lead to a new generation of technologies -- including, one day, cognitive prosthetics.
Unlocking patterns in Medical Imaging
An introduction to AI in medical imaging by Ben Glocker, Reader in Machine Learning for Imaging
Behind the recent victory of an artificial intelligence against a world champion player of Go lies the power of deep learning algorithms to compete in a broad range of application areas. Visual perception is one area where humans excel, but what if a machine could match or even exceed our ability to discriminate objects and identify patterns in what we see?
Hybrid human and machine learning
An introduction to hybrid human and machine learning by Emil Lupu, Professor of Computer Systems
Emil Lupu builds computer systems that can be continuously adapted to their environment and changes in how they are used. In his outlook for big data, Emil sees a world in which humans and machines collaborate to make sense of the huge amounts of unstructured data we are starting to gather from sensors.
Smart Dust
An introduction to Smart Dust by Julie McCann, Professor of Computer Systems
Professor Julie McCann takes her inspiration from nature for writing algorithms that allow distributed wireless networks to become self-organising, adaptive and intelligent.
Julie's technologies already find a home in monitoring water networks and to monitor pollution in cities. Looking ahead, sensors the size of dust will supersede today's cloud-based processing - computing intelligence will move out on the edge.
Human-computer partnership in the world of Big Data
An introduction to human-computer partnership by Nick Jennings, Professor of Artificial Intelligence
Humans have always been technology users but, in the era of big data and information overload, there is a need for more human-technology interaction. Through artificial intelligence some of these interactions can be explained, particularly with isolated and narrowly defined tasks. But the evolution of supercomputers, together with advances in machine learning offer extended opportunities for human-machine interaction. What if machines could respond to complex human commands? What if machine learning could determine when to ask humans or other artificial intelligence systems for help? What if artificial intelligence made it possible to model the unknown? Imperial College is leading in the use of supercomputing applications for human-machine collaboration.
Graduate Student Research Projects
Understanding the Spoken World
Aiden Hogg talks about his research and the problem of speaker diarisation
Most people find it quite easy to figure out who is talking, and when, just through hearing. However, computers find this very difficult. Aiden Hogg takes us through how computers do this now, and how his research makes allows computers to do this more easily.
Leaning-Augmented Algorithms to the Rescue
Ali Hadian describes the challenges around working with big data so that it can be searched quickly
Ali Hadian describes the challenges around how to work with big data so that it can be searched quickly.
How Robots See: The Mystery of the Convolutional Filter
Charlie Houseago teaches us about one of the most powerful tools available for computer vision
Charlie Houseago teaches us about one of the most powerful tools available for computer vision: the convolutional filter.
Metaprogrammes for Heterogeneous Systems
Jessica Vandebon tells us how we can effectively use heterogeneous systems to improve performance
To enable fast and energy efficient computing, modern large-scale computer systems contain specialist hardware that can accelerate parts of an application. However, programming these devices currently requires specialist skills that few programmers do have. Jessica Vandebon walks us through her research that aims to automatically apply optimisations for these heterogeneous systems to help programmers effectively use these.
Development of molecular-based point-of-care diagnostics
Kenny Malpartida Cárdenas tells us about her research in molecular-based point-of-care diagnostics
Kenny Malpartida Cárdenas talks about her research to develop tests for the detection antimicrobial resistance and infections diseases (including malaria and covid) that can be used without needing lab equipment.
Security in Cyber Physical Systems
Luca Castiglione talks about the importance of security in cyber physical systems
Luca Castiglione talks about the importance of security in cyber physical systems.
Can we guarantee AI is safe?
Michael Akintunde describes the importance of verification of automated systems
Michael Akintunde describes the importance of verification of automated systems and how researchers are attempting to achieve this verification.
Deep Learning on Custom Devices
Ruizhe Zhao talks us through his research, and how computers use convolutional neural networks
To perform visual tasks, like detecting objects in videos, computers can use "Convolutional Neural Networks". However, these can require lots of memory and be slow to execute. Ruizhe Zhao takes us through his research, which aims to effectively reduce the size of these networks to improve the performance of these networks.
DoC Clock
DoC Clock: Meet Konstantinos
Meet Konstantinos Barmpas, a PhD student from Imperial College London
Find out about Konstantinos' research in this special series of DoC Clock.
DoC Clock is the Department of Computing's new video series where staff and students have five minutes to discuss their research topics and personal interests.
(DoC is short for Department of Computing)
DoC Clock: Meet Margherita and Miguel
Meet Margherita and Miguel, two PhD students from the BioMedIA group
Find out about Margharita and Miguel's research in this special series of DoC Clock.
DoC Clock is the Department of Computing's new video series where staff and students have five minutes to interview each other and discuss their research topics and personal interests.
(DoC is short for Department of Computing)
DoC Clock: Meet Norman Di Palo
Meet Norman Di Palo, a PhD student from Imperial College London'd Robot Learning Lab
Find out about Norman's research in this special series of DoC Clock.
DoC Clock is the Department of Computing's new video series where staff and students have five minutes to discuss their research topics and personal interests.
(DoC is short for Department of Computing)
Contact
If you would like to know more about our outreach and public engagement,
email doc-outreach@imperial.ac.uk.