Sotabase

Career

· Entrepreneur In Residence, Cambridge Innovation Capital2025–2025
· Founder, Reasonable (Current)2025–
· Member Of The Scientific Committee, International Olympiad in Artificial Intelligence (IOAI)2024–2025
· Professor Of Machine Learning, University of Cambridge2024–
· Founder, Nagymaros AI Retreat (Current)2022–
· Project Founder, Chair Of Advisory Board, Ukraine Math and Science Achievement Fund2022–
· Associate Professor In Machine Learning, University of Cambridge2020–2024
· Honorary Lecturer, The Gatsby Unit, UCL2019–2024
· Principal Research Scientist, Magic Pony Technology (acquired by Twitter)2016–2016
· Senior Machine Learning Researcher, Twitter2016–2020
· Data Scientist, Balderton Capital2014–2015
· Visiting Researcher, Max Planck Institute for Intelligent Systems2011–2011
· Senior Data Scientist, PeerIndex (acquired by Brandwatch)2011–2014
· Visiting Researcher, Department of Engineering, University of Cambridge2009–2009
· Research Intern, Yahoo! Web Analytics2009–2009
· Summer Vacation Scholar, Department of Statistics, University of Oxford2008–2008
· Researcher, Sortware Engineer, Collegium Budapest2005–2008

Publications (35)

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
Computer Vision and Pattern Recognition · 2016
11,733
cited
Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network
Computer Vision and Pattern Recognition · 2016
5,746
cited
Lossy Image Compression with Compressive Autoencoders
International Conference on Learning Representations · 2017
1,121
cited
Bayesian Active Learning for Classification and Preference Learning
arXiv.org · 2011
1,022
cited
Amortised MAP Inference for Image Super-resolution
International Conference on Learning Representations · 2016
444
cited
How (not) to Train your Generative Model: Scheduled Sampling, Likelihood, Adversary?
arXiv.org · 2015
304
cited
Faster gaze prediction with dense networks and Fisher pruning
arXiv.org · 2018
224
cited
Is the deconvolution layer the same as a convolutional layer?
arXiv.org · 2016
147
cited
Variational Inference using Implicit Distributions
arXiv.org · 2017
141
cited
Collaborative Gaussian Processes for Preference Learning
Neural Information Processing Systems · 2012
130
cited
Note on the quadratic penalties in elastic weight consolidation
Proceedings of the National Academy of Sciences of the United States of America · 2017
119
cited
Optimally-Weighted Herding is Bayesian Quadrature
Conference on Uncertainty in Artificial Intelligence · 2012
111
cited
Approximate inference for the loss-calibrated Bayesian
International Conference on Artificial Intelligence and Statistics · 2011
75
cited
Model Size Reduction Using Frequency Based Double Hashing for Recommender Systems
ACM Conference on Recommender Systems · 2020
60
cited
Addressing delayed feedback for continuous training with neural networks in CTR prediction
ACM Conference on Recommender Systems · 2019
58
cited
Cognitive Tomography Reveals Complex, Task-Independent Mental Representations
Current Biology · 2013
58
cited
Deep Bayesian Bandits: Exploring in Online Personalized Recommendations
ACM Conference on Recommender Systems · 2020
44
cited
BRUNO: A Deep Recurrent Model for Exchangeable Data
Neural Information Processing Systems · 2018
35
cited
Jacobian-based Causal Discovery with Nonlinear ICA
Trans. Mach. Learn. Res. · 2023
27
cited
Efficient Wasserstein Natural Gradients for Reinforcement Learning
International Conference on Learning Representations · 2020
22
cited
Sotabase
Ferenc Huszar | Researcher Profile | Sotabase | Sotabase