Imperial College London

MrLancelotDa Costa

Faculty of Natural SciencesDepartment of Mathematics

Research Assistant



+44 (0)7908 414 831l.da-costa




Huxley BuildingSouth Kensington Campus





Biography. I am a Research Assistant and PhD student in the newly formed CDT Mathematics of Random Systems, jointly based at Imperial College London and University College London, under the co-supervision of Profs. Greg Pavliotis and Karl Friston. Prior to my PhD studies, I completed a Master in Pure Mathematics at the University of Cambridge and a Master of research in Brain Sciences at University College London. I did my undergraduate in mathematics at EPFL with a year on exchange at the University of Toronto.

Research. I am interested in the general principles of intelligent behaviour in biological and artificial agents. I research the Physics of Intelligent Systems by seeing how high-level descriptions of intelligent behaviour emerge from fundamental descriptions of matter. Complementarily, I apply theoretical insights to improve artificial agents. I spend most of my time learning about stochastic processes, information geometry, stochastic thermodynamics, neuroscience, and machine learning.

Publications. For an up to date list of publications see my Google scholar profile

Acknowledgements. I am grateful to my supervisors Professors Greg Pavliotis and Karl Friston, and to the National Research Fund in Luxembourg and EPSRC for supporting my PhD.



Da Costa L, Friston K, Heins C, et al., 2021, Bayesian mechanics for stationary processes, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol:477, ISSN:1364-5021, Pages:1-26

Friston K, Da Costa L, Hafner D, et al., 2021, Sophisticated Inference, Neural Computation, Vol:33, ISSN:0899-7667, Pages:713-763

Da Costa L, Parr T, Sajid N, et al., 2020, Active inference on discrete state-spaces: A synthesis, Journal of Mathematical Psychology, Vol:99, ISSN:0022-2496

Rubin S, Parr T, Da Costa L, et al., 2020, Future climates: Markov blankets and active inference in the biosphere, Journal of the Royal Society Interface, Vol:17, ISSN:1742-5689

More Publications