This job might no longer be available.
Research Internship, ABM Calibration
London, United Kingdom
4 months ago
Improbable believes in a future where new, virtual worlds will augment human experience and become as meaningful, lasting and rich as the physical world. We call this the Multiversal Self.
Our platform, SpatialOS, lets developers transcend the limits of regular computation, allowing swarms of servers running in the cloud to cooperate in order to simulate worlds far larger and more complex than any single server could.
At Improbable, you are surrounded by people who want to improve everything and everyone around them, and who compel you to improve yourself. We’re motivated by the fulfilment of solving hard problems to achieve something profound and transformative.
is to join our Research group, completing a 3-month PhD internship focussed on developing and assessing the efficacy of techniques for the calibration of agent-based models; using statistical inference, machine learning and optimisation. This internship is focused on the challenging problem of calibration for agent-based models (ABMs).
ABMs model social or biological systems at the level of individual entities interacting. For example, a crowd can be modelled as a large number of individual entities, each pursuing the goal of reaching a destination whilst avoiding collision with others. ABMs are a natural choice for modelling complex real-world systems. This complexity, however, can make it challenging to tune their parameters against real-world data. As a result we are investing long-term research efforts into scalable, effective tuning methods.
Areas for Impact
- You will be working with leading research with a preference towards statistically-inspired methods for tuning complex algorithms to real-world data. We are thinking about a combination of statistical inference techniques, like particle filters, statistical machine learning techniques, like random forests, and optimisation tools, like evolutionary programming.
- You will develop your expertise in one or more of these techniques, apply them to a specific ABM of a social system, and systematically study their efficacy.
- You will be encouraged to use a higher-level scientific computing language like Julia, R or Python, and we envisage that you will become adept at scientific programming (matrix algebra and probability).
- You’ll be joining a small team of experienced researchers backed by a large team of software engineers working on groundbreaking technology for national security.
- You will develop an understanding of the state of the art in the field and have the opportunity to advance it with support from other researchers, led by https://www.linkedin.com/in/canagnos/
We'd like to hear from you if you identify with the following
- Those currently pursuing a PhD in statistics, mathematics or a related area, and willing to take a 3-month interruption of studies for work experience.
- Those students with a background in physics or computer science that have had some training in statistics and/or optimisation are also welcome to apply.
- Expertise in scientific programming in a higher-level language for rapid prototyping. For example, statistical and scientific computing in Python, R, or Julia. Adept scientific programmers in other languages are still encouraged to talk to us, however.
- Although supervision will be provided, the ability to independently implement an algorithm (e.g. a particle filter) from a mathematical description is expected.
While we think the above experience could be important, we can’t predict the future and so we’re keen to hear from applicants that believe they have valuable experience. If you identify with the team & mission, but not all of the suggestions, then please still apply!!
The best ideas are often the least expected and require new ways of thinking; that’s why our teams at Improbable are made up of an incredible range of talented people. Improbable is proud to be an equal opportunity employer. We do not discriminate based on race, ethnicity, colour, ancestry, national origin, religion, sex, sexual orientation, gender identity, age, disability, veteran status, genetic information, marital status or any other legally protected status.