"Warning, in music-words
devout and large,
that we are each other’s
we are each other’s
we are each other’s
magnitude and bond."

Gwendolyn Brooks, 1971

Core Members

"The source of these problems of coordination and cooperation is not the nature of the individuals’ goals, or the instrumental character of rationality. Rather it is individualism about rationality, which holds the unit of activity exogenously fixed at the individual."
Hurley 2004

Eduardo Alonso

sample-image I am a Reader in Computing at the Artificial Intelligence Research Centre, where I study relationships between reinforcement learning and optimization techniques, associative learning, and evolutionary models of aposematism and foraging. I have a keen interest in exploring variational principles and symmetries in learning and behaviour.

More generally, I am interested in applications of AI technology to real-life problems, such as solutions for energy, health, and biochemistry, and in the ethical and legal impact of deep learning, and in investigating whether AI algorithms can produce creative art.

I have published dozens of papers in journals like Neural Computation, Neural Networks, IEEE TNN&LS, and have contributed to The Cambridge Handbook of Artificial Intelligence and several Springer's LNAI and LNCS volumes. My work has been spotlighted by the IEEE Computational Intelligence Society as one of the most important contributions in the area in 2013, and was awarded the First Prize of the European Institute of Innovation and Technology ICT 2014.

I have acted as vice-chair of The Society for the Study of Artificial Intelligence and the Simulation of Behaviour (AISB), the eldest learned Artificial Intelligence society, and am a member of the EPSRC College.

Esther Mondragón

sample-image I am a lecturer at the Artificial Intelligence Research Centre and a computational cognitive neuroscientist working in nature-inspired AI. My main research focusses on modelling associative learning (AL), and in the conceptual and formal modifications in learning theory that would allow incorporating phenomena that at face value are not susceptible to associative analysis. I have developed several computational learning models such as the Double error Dynamic Asymptote (DDA) model, a fully-connected architecture for Pavlovian conditioning with a dynamic asymptote, which determines the direction of learning, and the Rescorla-Wagner Drift-Diffusion Model (RWDDM) that combines a noisy linear accumulator and the Rescorla-Wagner learning rule.

My current research centres on integrating deep learning architectures and associative learning to explore the possible contribution of associative processes to the formation of representational hierarchies, as a first necessary step towards scaling up to higher-order cognition. AL is core to bottom-up approaches to natural intelligence and thus keystone to the development of AI algorithms and architectures meant to simulate human-like behaviour.

I have published in presetigious journals such as Psychological Review, Trends in Cognitive Science, PloS Computational Biology, and Science.

I am a senior member of The Society for the Study of Artificial Intelligence and the Simulation of Behaviour (AISB)

Laure Daviaud

sample-image I am a lecturer at City, University of London and member of CitAI. Prior to this, I had the luck to work with amazing researchers: I did my PhD at the University Paris VII under the supervision of Thomas Colcombet and Jean-Eric Pin, and spent a few years as a post-doctoral research fellow in Aix-Marseille University (with Pierre-Alain Reynier and Jean-Marc Talbot), ENS Lyon (with Colin Riba), University of Warsaw (with Mikolaj Bojanczyk) and the University of Warwick (with Marcin Jurdzinski and Ranko Lazic).

I am studying mathematical abstractions for verification problems, such as automata, games and logic formalisms. This way, I try to understand the limitations of computational models, how powerful they are, how fast …

I have studied more precisely optimization models (max-plus automata and tropical algebra), transducers and games (parity and mean-payoff) for example.

Alex Ter-Sarkisov

sample-image I am lecturer in the Computer Science Department at City, University of London.

My recent work is mostly on Deep Learning in Computer Vision (instance segmentation), but I'm also interested in Genetic Algorithms, both theory (mathematical analysis of runtime) and applications (new genetic operators). I like teaching artificial intelligence, both at undergraduate and advanced (ConvNets, FCNs, object detection/segmentation) levels.

I obtained my PhD at Massey University, New Zealand, in 2012.

I worked as postdoctoral researcher at the University of Waterloo (2012–2013), the Universite du Maine (2014–2015), and at the Dublin Institute of Technology (2015 – 2018).

Michaël Garcia-Ortiz

sample-image I obtained my PhD from Bielefeld University in 2013. The doctoral work, on the topic of ‘Prediction of Driver Behavior’, was performed in collaboration with Honda Research Institute Europe, and involved the prediction of driver behaviors for Advanced Driving Assistant Systems. This work was followed by a PostDoc in Ensta ParisTech, on the topic of pedestrian detection and tracking. I joined Softbank Robotics Europe (formerly Aldebaram), a Paris based robotics company, in September 2013.

For more than 5 years, I was a Research Scientist in SBRE’s AI Lab, where I conducted research on the topic of Artificial General Intelligence. I participated there in the European Project APRIL, focused on learning for personal robotics.

I joined City in July 2019 as a Lecturer for the new Masters in AI. I am a member of the new Artificial Intelligence Research Center CitAI, where I continue my research on AGI and the emergence of common sense knowledge.

Giacomo Tarroni

sample-image I have been a full-time researcher in the field of medical image analysis since 2009. My work has been mainly focused on image segmentation, image registration, quality control and object tracking for cardiovascular, brain and fetal images. In particular, I obtained my Ph.D. from the University of Bologna, Italy (in collaboration with the University of Chicago, U.S.) working on the automated analysis of first-pass myocardial perfusion sequences in MRI. During my post-doc at the University of Padova, Italy, I focused on the automated analysis of fetal ultrasound images. After being awarded a Marie Skłodowska-Curie Fellowship from the European Commission, I moved to Imperial College London, where I became interested in the applications of machine learning and AI to automated organ detection, quality control assessment and motion correction for cardiac MRI.

My current research focus is on artificial intelligence approaches for image quality assessment through outlier detection, shape analysis through representation learning and image-to-image translation through generative models, both for medical image analysis and more generally for computer vision.

Alex Galkin

sample-image I am a Teaching Assistant in Data Science at City, University of London for modules such as Machine Learning, Principles of Data Science, Visual Analytics, Neural Computing, Computer Vision and Big Data.

In my research I work on speech separation problems using CNN and Capsule Networks.

My current main interest is employing deep reinforcement learning approaches for NLP problems as well as applying Minimum Hyperspherical Energy (MHE) for neural networks to improve their generalisation ability in various applications.

Recent works:

Staines, T.; Weyde, T. and Galkin, O. "Monaural Speech Separation with Deep Learning Using Phase Modelling and Capsule Networks", 27th European Signal Processing Conference (EUSIPCO 2019), A Coruña, Spain

Perez-Lapillo, J.; Galkin, O and Weyde, T. "Improving Singing Voice Separation With The Wave-U-Net Using Minimum Hyperspherical Energy" (submitted), 45th International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020, Barcelona, Spain