School of Informatics
University of Edinburgh
10 Crichton Street
+44 131 651 3091
Mobile: +44 7879 671 811
Secretary: +44 131 650 3100
Fax: +44 870 130
University of Edinburgh Summary Page
My research focus in the last few years has moved from computational
neuroscience and neuroinformatics to the application of
machine-learning methods in energy-related areas. I also have a
subsidiary interest in developing new approaches to macroeconomics
based on accounting of real flows in the economy.
- IDEAL: Intelligent Domestic Energy Advice Loop.
Reducing energy demand from existing dwellings through occupant
behaviour change is crucial for meeting UK carbon emission reduction
targets and will help the increasing numbers of households struggling
to pay their energy bills. This project explores the
interaction of energy technologies and householder behaviours related
to energy in a sample of 600 dwellings. Using wireless sensing,
machine learning, and natural language generation technologies, we
will construct an intelligent advice loop that will provide
information to householders on what activities they are engaging in
which use energy and how much energy is used for each one, together
with suggestions for what they might do to reduce their energy
expenditure; for example: "Last week you spent £10 on hot water for
showers, if you reduced your average shower time from 15 minutes to 12
minutes you could save £100 per year". If we can show that this loop
is effective in helping people to reduce their energy demand, then we
expect that energy suppliers and other companies will start to offer
it as a service to households to help them keep their energy costs
I am Principal Investigator on this project and have most interest in
the machine learning methods we apply to infer behavours from noisy
sensor data. I also oversee the software engineering aspects of the
data pipeline and the feedback channels.
This project studies the extent to which national surveys of energy use in homes can be carried out remotely and largely automatically using smart meter data augmented with carefully targeted additional sensor data. It is a spin-off from the IDEAL project, and is a collaboration with UCL's Centre for Energy Epidemiology.
Enhance is a research project which is taking a Living Lab approach to understanding and reducing energy demand in public sector buildings. The project is a collaboration between the School of Informatics, School of Social and Political Science, and Edinburgh College of Art at the University of Edinburgh. I am Principle Investigator.
- 7see macroeconomic model
This project, led by Simon Roberts at Arup, is a multidisciplinary
collaboration aimed at understanding the long-term drivers and
constraints on evolution of the macroeconomy, in the tradition of
biophysical economics. We use system dynamics modelling methods
together with extensive analysis of national macroeconomic data to
enable evaluation of policy options related to natural resources
(principally energy). Two versions of the UK model have been
released (here and here) . The fundamental theses of this work is that, in contrast to
standard macroeconomic models:
- There are serious constraints on substitutability in the short
and medium term. An obvious example is the forty-year commitment
to energy sources.
- The system does not tend to an equilibrium but is a complex
chaotic system that nevertheless in the right circumstances
exhibits stability and strong patterns (think of weather and
- Energy is a fundamental enabler of economic activity, and in a
time of uncertain energy supplies can act as a significant
constraint on economic activity.
I currently have one Ph.D. student and am open to taking on more
students who are interested in working in areas related to my
- Sergio Elizondo Gonzalez is working on a combination of machine-learning and agent-based methods to address the balance of supply and demand in future electricity networks (the Smart Grid).
I am Co-Director
Environment and Society
, which brings together Ph.D. students across the
University who are working on interdisciplinary projects within the
information for PhD
Other ActivitiesI was a member of the EU Smart Cities Initiative Stakeholder Platform,
in the Energy Efficiency and Buildings Working Group. Within this
group, I was Topic Coordinator for the "Living Labs and the Smart City"
topic. We developed a set of proposals for the European
Commission to advise them on key innovations to focus on over the
period to 2020.
- Introductory Applied Machine Learning This is our entry-level course into machine learning, for 3rd year, 4th year and MSc students, on campus and remote. It runs on a flipped model and I use it to study how university teaching can be done at scale.
From September 2002 to July 2006 I
was Founding CEO of the business that has become ResearchSpace Ltd, which provides an online Electronic Lab Notebook. I continue to advise and support the company.