Paul Goldberg
Paul Goldberg
Room
255,
Wolfson Building,
Parks Road, Oxford OX1 3QD
United Kingdom
Interests
I’m interested in hearing from prospective research students or other collaborators. Topics/key phrases:
- machine learning; game-theoretic models for data
- Agent-based models; calibration with real-world data
- decentralized computation of economic equilibria
- computational complexity, communication complexity, query complexity
personal web page (with link to list of my papers, current teaching, etc) (tends to be more detailed and up-to-date than this page.)
Algorithms and Data Structures Moodle
Biography
Since July 2013 I have been a professor at the Department of Computer Science, Oxford University.
Prior to that I was a professor of Computer Science at the University of Liverpool, where I was founding head of the Economics and Computation (ECCO) research group. I have also taught at the University of Warwick, and been a member of research groups at Aston University and Sandia National Labs, USA.
The unifying theme of my work is algorithms having mathematically proven performance guarantees. I have worked extensively in computational learning theory (machine learning algorithms, sample size bounds). Some other work is in computational biology and approximation algorithms. Most of my recent work since 2002 is in algorithmic game theory (complexity of equilibrium computation, analysis of best-response and better-response dynamics). More recently (since 2017) I have also worked on computational complexity of "total search problems" in the complexity class NP (this stems from earlier work on the complexity of Nash equilibrium computation).
Activities
- Foundational Issues in Computational Learning
- Algorithms At Large
- Algorithmic Game Theory and Computational Economics