Per Kristian Lehre (@pklehre) 's Twitter Profile
Per Kristian Lehre

@pklehre

Professor, Computer Science, University of Birmingham (UK). Theory of #EvolutionaryComputation @[email protected]

ID: 406529308

linkhttp://www.cs.bham.ac.uk/~lehrepk calendar_today06-11-2011 20:47:59

943 Tweet

902 Followers

1,1K Following

Per Kristian Lehre (@pklehre) 's Twitter Profile Photo

Learning in games usually assumes small action spaces. This afternoon at #AAAI2025 we give an oral presentation showing that the PDCoEA co-evolutionary algorithm finds the Nash Equilibrium of the game below (2^n actions) in expected poly(n) time. Joint work with @lss1242

Learning in games usually assumes small action spaces. This afternoon at #AAAI2025 we give an oral presentation showing that the PDCoEA co-evolutionary algorithm finds the Nash Equilibrium of the game below (2^n actions) in expected poly(n) time. Joint work with @lss1242
Per Kristian Lehre (@pklehre) 's Twitter Profile Photo

Luca Manzoni presenting work by Nils Aall Barricelli in the 1950s. Perhaps the first experiments with a coevolutionary algorithm.

Luca Manzoni presenting work by Nils Aall Barricelli in the 1950s. Perhaps the first experiments with a coevolutionary algorithm.
Per Kristian Lehre (@pklehre) 's Twitter Profile Photo

Evolutionary Algorithms continue to make strong impact in AI and machine learning, particularly when combined with LLMs. This latest work by Google Deepmind goes beyond FunSearch.

Per Kristian Lehre (@pklehre) 's Twitter Profile Photo

Interesting talk by Papadimitrou at the Nash@75 workshop in Oxford today. He proposed an alternative more computationally feasible solution concept for games based on replicator dynamics. However, I think the infinite population approximation is unrealistic for large games.

Interesting talk by Papadimitrou at the Nash@75 workshop in Oxford today. He proposed an alternative more computationally feasible solution concept for games based on replicator dynamics. However, I think the infinite population approximation is unrealistic for large games.
Per Kristian Lehre (@pklehre) 's Twitter Profile Photo

Alistair Benford presenting our runtime analysis of co-evolutionary algorithms on combinatorial games at GECCO 2026 He shows that the runtime can be upper bounded in terms of a graph invariant called switchqbility.

Alistair Benford presenting our runtime analysis of co-evolutionary algorithms on combinatorial games at <a href="/GeccoConf/">GECCO 2026</a> He shows that the runtime can be upper bounded in terms of a graph invariant called switchqbility.
Per Kristian Lehre (@pklehre) 's Twitter Profile Photo

Happy to join as Associate Editor of ACM Transactions on Evolutionary Learning and Optimization. The journal covers an extremely exciting area at the intersection of Machine Learning and Optimization. dl.acm.org/journal/telo

UoB Computer Science (@uobcompsci) 's Twitter Profile Photo

This is so exciting! We're happy to be a partner university with Martingale Foundation. If you are a home student considering studying our MSc Artificial Intelligence and Machine Learning, discover more about this amazing scholarship here: martingale.foundation/partner-univer…