About Me

I am a third year PhD student in the machine learning group supervised by Prof Zoubin Ghahramani, advised by Prof Carl E. Rasmussen, and funded by the Google Europe Doctoral Fellowship. Before arriving at Cambridge I studied at the Computer Science department at Oxford for a Master's degree under the supervision of Prof Phil Blunsom. In my research at Oxford I worked on machine learning and natural language processing, and in particular on machine translation and the development of new probabilistic word alignment models. Prior to my MSc I worked as a software engineer for 3 years at IDesia Biometrics developing code and UI for mobile platforms. I did my undergraduate in mathematics and computer science at the Open University in Israel, where I worked on computational biology, deep learning, and game theory as a fourth year project.


New blog post and more


I added a new blog post about deep learning and scientific methodology. Also added new papers [1, 2, 3] and NIPS posters [1, 2, 3], as well as ICML videos [1, 2] and a video of the talk ``Modern Deep Learning through Bayesian Eyes'' given at Microsoft Research.

``Modern Deep Learning through Bayesian Eyes'' slides online


Talks page has been updated with a talk given at UCL as part of the ``Computational Statistics and Machine Learning'' seminar series.

Invited talks


I'll be talking about Deep Learning Uncertainty at UCL on 13/11, ETH (Zurich) on 16/11, ATI Deep Learning workshop (Edinburgh) on 25/11, and Microsoft Research (Cambridge) on 26/11.

Invited Presentation


I will present Dropout as a Bayesian Approximation at the ``First Deep Learning Symposium'' at NIPS 2015.

New extrapolated art works


extrapolated art has been updated with new art works.

Invited talk


Talks page has been updated with an invited talk given at ``Bayesian nonparametrics in the North'' meeting, Ecole Centrale de Lille, 2015.

ICML updates


Presentations and posters for the 2 ICML papers [1] [2] have been added as well as for the talk at the 10th Conference on Bayesian Nonparametrics.

Qualcomm Innovation Fellowship


Many thanks to Qualcomm for the Qualcomm Innovation Fellowship award.

Research interests

My interests lie in the fields of linguistics, applied maths, and computer science. Most of my work is motivated by problems found in the intersections of these fields. In natural language processing my current research covers generative language models defined over continuous spaces, scalable models for compositional semantics, and adaptor grammars; In maths I worked on fragmentation processes, improper Poisson-Dirichlet distributions, and the application of ergodic theory to probabilistic modelling; I'm currently working on probabilistic models of combinatorial structures such as the ones underlying multi-view clustering, and distribution estimation. In computer science and machine learning I work on distributed algorithms for inference in non-parametric models such as the Dirichlet process and Gaussian process, and study different computational representations of semantics. A list of publications is available here.

I am a member of Gonville and Caius College.

Curriculum vitae

My Résumé is available here.

Contact me


yg279 -at- cam.ac.uk


Cambridge University
Engineering Department
Cambridge, CB2 1PZ
United Kingdom