skip to content

Statistical Laboratory

Ioannis Kontoyiannis is the Churchill Professor of Mathematics of Information in DPMMS, and a Fellow of Darwin College. Currently he is Chair of the Faculty of Mathematics. 

Professor Kontoyiannis works in information theory, applied probability, and statistics, including their applications in neuroscience, bioinformatics, and the development of machine learning algorithms. His work with collaborators includes foundational results in entropy estimation, the spectral theory of Markov processes, the fundamental limits of data compression and communication, and information-theoretic methods for time-series modelling. Recent projects concern the sample complexity of compression and channel coding, entropy inequalities and additive combinatorics, Bayesian causal discovery, and stochastic algorithms for optimisation and learning.

His research has been funded by EPSRC, the National Science Foundation, the European Union, Greek national funds, the European Research Council, and numerous other national and international bodies. He has also been involved in consulting work for the Government and for companies in the financial, medical, and high-tech industries. He is currently a co-investigator with EPSRC's INFORMED-AI Hub working on information-theoretic foundations of distributed AI. 

Kontoyiannis was born in Athens, Greece, in 1972. He studied mathematics at Imperial College and Cambridge, and received the Ph.D. degree in electrical engineering from Stanford University. He has taught at Brown, Columbia, Purdue, and the Athens University of Economics and Business. Between 2018 and 2020 he was Professor of Information and Communications with the Information Engineering Division of the Engineering Department at Cambridge, where he was also Head of the Signal Processing and Communications Laboratory, and where he remains as an affiliated member. He has been with DPMMS since June 2020.

He has been awarded the Manning Endowed Assistant Professorship and an Honorary Master of Arts Degree Ad Eundem by Brown University; he has held Sloan Foundation and Marie Curie Fellowships. He is a Fellow of the IEEE and of the IMS.

Further information about his research, publications and professional activities is available on his personal website.

 

Publications

Conditions for equality and stability in Shannon’s and Tao’s entropy power inequalities
L Gavalakis, I Kontoyiannis
– IEEE Transactions on Information Theory
(2026)
PP,
1
Fine asymptotics for approximate simulation of symmetric channels
V Lungu, AB Wagner, I Kontoyiannis
(2026)
Fast PAC Global Optimization via Restarted Langevin: Exploration, Exploitation, and Degenerate Cooling
I Kontoyiannis, S Meyn
(2026)
Shakespeare, Entropy and Educated Monkeys
I Kontoyiannis
– IEEE BITS the Information Theory Magazine
(2026)
PP,
1
ConRAD: Conformal Risk-Aware Neural Databases
S Horchidan, F Zeiher, X Shi, V Kalavri, H Boström, I Kontoyiannis, P Carbone
– Proceedings of the VLDB Endowment
(2026)
19,
3511
The Sample Complexity of Lossless Data Compression
T Viaud, I Kontoyiannis
– IEEE Transactions on Information Theory
(2026)
72,
8177
The Sample Complexity of Lossless Data Compression
T Viaud, I Kontoyiannis
(2026)
Entropy lower bounds and sum-product phenomena
L Gavalakis, MK Goh, I Kontoyiannis
(2026)
Entropy power inequalities in compact groups
L Gavalakis, I Kontoyiannis, SM Sriramu, A Wagner
(2026)
Sample Complexity Bounds for Lossless Source Coding
T Viaud, I Kontoyiannis
– 2026 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, ISIT
(2026)
00,
1
  • 1 of 23
  • >

Frontpage talks

Cambridge Statistics Clinic

Statistics

Statistics

Cambridge Statistics Clinic

Research Group

Statistical Laboratory

Room

D1.09

Telephone

01223 337961