PhD Scholarship: Visual explanations of type inference
University of Leeds
|Funding for:||UK Students, EU Students, International Students|
|Funding amount:||Maintenance: £14,296|
|Placed on:||1st December 2016|
|Closes:||6th January 2017|
Name of School Contact: David Duke
Contact Details: Tel: +44 113 343 6800
How to apply: http://www.leeds.ac.uk/info/130206/applying/91/applying_for_research_degrees
Degree Level: Research Postgraduate
Scholarship Type: International Home/EU
Number available: 1
Funding Type: School of Computing
Applicants should read the information below for further details on the scholarship or visit the School web site given below.
The functional programming language Haskell is a "living laboratory" for innovation in data types and type inference, with features such as GADTs, associated types, type constraints, and type families enhancing the expressive power and allowing the compiler to reject more "bad" programs. However, the increased richness makes it increasingly difficult to discover the underlying cause of type errors, particularly when using sophisticated libraries. Reporting of type errors is improving, but still relies on presenting the programmer with a textual "snapshot" of the information available to the type inference engine at the point where an error is detected. The objective of this project is to take a fresh look at how decisions made in type inference, and in particular the reporting of type errors, are presented to the programmer. The starting point will be looking at models of diagnostic reasoning, and the use of information visualization to display the process that lead to a conclusion in a way that is much more concise and usable than a simple verbose listing of inference rules.
Skills needed for the project include an understanding of type systems and type inference, and strong programming skills in a functional language (ideally Haskell). Software development including graphics will be done in Haskell, and may require interfacing with the type inference engine within the GHC compiler. Prior exposure to information visualization would also be an advantage.
This award will cover the cost of academic fees and provide a maintenance allowance of at least £14,296 per year for a maximum for 3 years, subject to satisfactory progress.
Minimum Academic Requirements (if English is not your first language, then candidates must also meet the University’s English language requirements):
Applications are invited from candidates who have achieved or expecting a minimum of a UK upper second class (hons) degree or equivalent, in a field relevant to the project.
If you first language is not English, you will need to satisfy the University of Leeds English language requirements: http://www.leeds.ac.uk/projectleeds/info/123100/admissions/143/entry_requirements
Share this PhD
Type / Role: