Freddie Bickford Smith

• Personal Webpage
• GitHub
• Twitter
• Google Scholar

Deep Learning, Uncertainty Estimation

I’m a DPhil student working with Tom Rainforth, Adam Foster and Jakob Foerster. My aim is to address shortfalls of deep learning using principled probabilistic methods. Currently I’m adapting ideas from Bayesian experimental design with a view to improve the data efficiency of neural networks and other predictive models. I have also worked on an approximate-inference scheme to allow neural networks to learn more effectively across sequences of tasks. Previously, during my MSc at UCL, I collaborated with Brad Love, Brett Roads and Ed Grefenstette on a project aimed at understanding neural networks from a cognitive-science perspective. As an undergraduate I studied mechanical engineering at Bristol, working on carbon-fibre composites with Fabrizio Scarpa and on 3D printing with Ben Hicks.

Publications

2023

• F. Bickford Smith , A. Kirsch , S. Farquhar , Y. Gal , A. Foster , T. Rainforth , Prediction-oriented Bayesian active learning, International Conference on Artificial Intelligence and Statistics, 2023.
  

  @article{bickfordsmith2023prediction,
    title = {Prediction-oriented {Bayesian} active learning},
    author = {Bickford Smith, Freddie and Kirsch, Andreas and Farquhar, Sebastian and Gal, Yarin and Foster, Adam and Rainforth, Tom},
    year = {2023},
    journal = {International Conference on Artificial Intelligence and Statistics}
  }
  

2022

• T. Rudner , F. Bickford Smith , Q. Feng , Y. W. Teh , Y. Gal , Continual learning via sequential function-space variational inference, International Conference on Machine Learning, 2022.
  

  @article{rudner2022continual,
    title = {Continual learning via sequential function-space variational inference},
    author = {Rudner, Tim and Bickford Smith, Freddie and Feng, Qixuan and Teh, Yee Whye and Gal, Yarin},
    year = {2022},
    journal = {International Conference on Machine Learning}
  }