Richard Connor

Richard Connor is a Professor of Computer Science at the University of St Andrews' School of Computer Science. He holds a BSc, PhD, and postdoctoral fellowships from the same institution, with prior roles including Director of Research and Head of Department at the University of Strathclyde. His research focuses on similarity search, high-dimensional geometry, and metric spaces, with contributions to algorithms like nSimplexZen and BitPart. He teaches distributed systems and advises PhD students, including Ben Claydon. Connor has engaged in tech startups and contributed to Scotland's computing education curriculum. His work addresses challenges in large-scale data search, leveraging geometric principles and innovative indexing techniques.

• Education: BSc and PhD from University of St Andrews (1981–1996)
• Key Roles: EPSRC Advanced Fellow, Chair at Strathclyde, Head of Division at Stirling, and return to St Andrews (2020)

Research Interests: Connor investigates similarity search in high-dimensional spaces, focusing on metric geometry, Hilbert spaces, and dimensionality reduction. His work bridges theoretical foundations and practical applications like image retrieval and machine learning optimization. Recent projects include polyadic queries and Ptolemaic partitioning mechanisms to enhance search efficiency.

Publications: His 2024 and 2025 works explore novel dimensionality reduction (nSimplexZen), cross-entropy loss correlations, and scalable polyadic queries, advancing the field through algorithmic innovation. Earlier contributions include BitPart for high-dimensional metric search and analysis of binary partitioning efficacy.

Awards: Not explicitly listed, but his EPSRC Advanced Fellowship and academic leadership roles highlight recognition in the field.

Advising & Grants: Supervised Ben Claydon’s PhD; led research funded via fellowships and institutional roles. Active in organizing academic events like Doors Open @ Computer Science.

Labs/Teams: Engaged with MetricSearch research groups and collaborates internationally on similarity search projects.