Timezone: »
Poster
How Tight Can PAC-Bayes be in the Small Data Regime?
Andrew Foong · Wessel Bruinsma · David Burt · Richard Turner
In this paper, we investigate the question: _Given a small number of datapoints, for example $N = 30$, how tight can PAC-Bayes and test set bounds be made?_ For such small datasets, test set bounds adversely affect generalisation performance by withholding data from the training procedure. In this setting, PAC-Bayes bounds are especially attractive, due to their ability to use all the data to simultaneously learn a posterior and bound its generalisation risk. We focus on the case of i.i.d. data with a bounded loss and consider the generic PAC-Bayes theorem of Germain et al. While their theorem is known to recover many existing PAC-Bayes bounds, it is unclear what the tightest bound derivable from their framework is. For a fixed learning algorithm and dataset, we show that the tightest possible bound coincides with a bound considered by Catoni; and, in the more natural case of distributions over datasets, we establish a lower bound on the best bound achievable in expectation. Interestingly, this lower bound recovers the Chernoff test set bound if the posterior is equal to the prior. Moreover, to illustrate how tight these bounds can be, we study synthetic one-dimensional classification tasks in which it is feasible to meta-learn both the prior and the form of the bound to numerically optimise for the tightest bounds possible. We find that in this simple, controlled scenario, PAC-Bayes bounds are competitive with comparable, commonly used Chernoff test set bounds. However, the sharpest test set bounds still lead to better guarantees on the generalisation error than the PAC-Bayes bounds we consider.
Author Information
Andrew Foong (University of Cambridge)
PhD student from Malaysia working on Bayesian deep learning at the Computational and Biological Learning lab at the Cambridge University Engineering Department.
Wessel Bruinsma (University of Cambridge and Invenia Labs)
David Burt (University of Cambridge)
Richard Turner (University of Cambridge)
More from the Same Authors
-
2022 : Ice Core Dating using Probabilistic Programming »
Aditya Ravuri · Tom Andersson · Ieva Kazlauskaite · William Tebbutt · Richard Turner · Scott Hosking · Neil Lawrence · Markus Kaiser -
2022 : Active Learning with Convolutional Gaussian Neural Processes for Environmental Sensor Placement »
Tom Andersson · Wessel Bruinsma · Efstratios Markou · Daniel C. Jones · Scott Hosking · James Requeima · Anna Vaughan · Anna-Louise Ellis · Matthew Lazzara · Richard Turner -
2022 : Contextual Squeeze-and-Excitation »
Massimiliano Patacchiola · John Bronskill · Aliaksandra Shysheya · Katja Hofmann · Sebastian Nowozin · Richard Turner -
2022 : FiT: Parameter Efficient Few-shot Transfer Learning »
Aliaksandra Shysheya · John Bronskill · Massimiliano Patacchiola · Sebastian Nowozin · Richard Turner -
2022 : Adversarial Attacks are a Surprisingly Strong Baseline for Poisoning Few-Shot Meta-Learners »
Elre Oldewage · John Bronskill · Richard Turner -
2023 Poster: PDE-Refiner: Achieving Accurate Long Rollouts with Neural PDE Solvers »
Phillip Lippe · Bas Veeling · Paris Perdikaris · Richard Turner · Johannes Brandstetter -
2023 Poster: Geometric Neural Diffusion Processes »
Emile Mathieu · Vincent Dutordoir · Michael Hutchinson · Valentin De Bortoli · Yee Whye Teh · Richard Turner -
2023 Poster: Kronecker-Factored Approximate Curvature for Modern Neural Network Architectures »
Runa Eschenhagen · Alexander Immer · Richard Turner · Frank Schneider · Philipp Hennig -
2023 Poster: Timewarp: Transferable Acceleration of Molecular Dynamics by Learning Time-Coarsened Dynamics »
Leon Klein · Andrew Foong · Tor Fjelde · Bruno Mlodozeniec · Marc Brockschmidt · Sebastian Nowozin · Frank Noe · Ryota Tomioka -
2022 : Panel »
Erin Grant · Richard Turner · Neil Houlsby · Priyanka Agrawal · Abhijeet Awasthi · Salomey Osei -
2022 Poster: Sparse Gaussian Process Hyperparameters: Optimize or Integrate? »
Vidhi Lalchand · Wessel Bruinsma · David Burt · Carl Edward Rasmussen -
2022 Poster: Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification »
Massimiliano Patacchiola · John Bronskill · Aliaksandra Shysheya · Katja Hofmann · Sebastian Nowozin · Richard Turner -
2021 Poster: Collapsed Variational Bounds for Bayesian Neural Networks »
Marcin Tomczak · Siddharth Swaroop · Andrew Foong · Richard Turner -
2021 Poster: Memory Efficient Meta-Learning with Large Images »
John Bronskill · Daniela Massiceti · Massimiliano Patacchiola · Katja Hofmann · Sebastian Nowozin · Richard Turner -
2020 Poster: Efficient Low Rank Gaussian Variational Inference for Neural Networks »
Marcin Tomczak · Siddharth Swaroop · Richard Turner -
2020 Poster: Meta-Learning Stationary Stochastic Process Prediction with Convolutional Neural Processes »
Andrew Foong · Wessel Bruinsma · Jonathan Gordon · Yann Dubois · James Requeima · Richard Turner -
2020 Poster: On the Expressiveness of Approximate Inference in Bayesian Neural Networks »
Andrew Foong · David Burt · Yingzhen Li · Richard Turner -
2020 Poster: VAEM: a Deep Generative Model for Heterogeneous Mixed Type Data »
Chao Ma · Sebastian Tschiatschek · Richard Turner · José Miguel Hernández-Lobato · Cheng Zhang -
2020 Poster: Continual Deep Learning by Functional Regularisation of Memorable Past »
Pingbo Pan · Siddharth Swaroop · Alexander Immer · Runa Eschenhagen · Richard Turner · Mohammad Emtiyaz Khan -
2020 Oral: Continual Deep Learning by Functional Regularisation of Memorable Past »
Pingbo Pan · Siddharth Swaroop · Alexander Immer · Runa Eschenhagen · Richard Turner · Mohammad Emtiyaz Khan -
2019 Poster: Icebreaker: Element-wise Efficient Information Acquisition with a Bayesian Deep Latent Gaussian Model »
Wenbo Gong · Sebastian Tschiatschek · Sebastian Nowozin · Richard Turner · José Miguel Hernández-Lobato · Cheng Zhang -
2019 Poster: Practical Deep Learning with Bayesian Principles »
Kazuki Osawa · Siddharth Swaroop · Mohammad Emtiyaz Khan · Anirudh Jain · Runa Eschenhagen · Richard Turner · Rio Yokota -
2018 Poster: Infinite-Horizon Gaussian Processes »
Arno Solin · James Hensman · Richard Turner -
2018 Poster: Geometrically Coupled Monte Carlo Sampling »
Mark Rowland · Krzysztof Choromanski · François Chalus · Aldo Pacchiano · Tamas Sarlos · Richard Turner · Adrian Weller -
2018 Spotlight: Geometrically Coupled Monte Carlo Sampling »
Mark Rowland · Krzysztof Choromanski · François Chalus · Aldo Pacchiano · Tamas Sarlos · Richard Turner · Adrian Weller -
2017 Poster: Streaming Sparse Gaussian Process Approximations »
Thang Bui · Cuong Nguyen · Richard Turner -
2017 Poster: Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning »
Shixiang (Shane) Gu · Timothy Lillicrap · Richard Turner · Zoubin Ghahramani · Bernhard Schölkopf · Sergey Levine -
2016 Poster: Rényi Divergence Variational Inference »
Yingzhen Li · Richard Turner -
2015 Poster: Neural Adaptive Sequential Monte Carlo »
Shixiang (Shane) Gu · Zoubin Ghahramani · Richard Turner -
2015 Poster: Learning Stationary Time Series using Gaussian Processes with Nonparametric Kernels »
Felipe Tobar · Thang Bui · Richard Turner -
2015 Poster: Stochastic Expectation Propagation »
Yingzhen Li · José Miguel Hernández-Lobato · Richard Turner -
2015 Spotlight: Learning Stationary Time Series using Gaussian Processes with Nonparametric Kernels »
Felipe Tobar · Thang Bui · Richard Turner -
2015 Spotlight: Stochastic Expectation Propagation »
Yingzhen Li · José Miguel Hernández-Lobato · Richard Turner -
2014 Poster: Tree-structured Gaussian Process Approximations »
Thang Bui · Richard Turner -
2014 Spotlight: Tree-structured Gaussian Process Approximations »
Thang Bui · Richard Turner -
2011 Poster: Probabilistic amplitude and frequency demodulation »
Richard Turner · Maneesh Sahani -
2011 Spotlight: Probabilistic amplitude and frequency demodulation »
Richard Turner · Maneesh Sahani -
2009 Poster: Occlusive Components Analysis »
Jörg Lücke · Richard Turner · Maneesh Sahani · Marc Henniges -
2007 Workshop: Beyond Simple Cells: Probabilistic Models for Visual Cortical Processing »
Richard Turner · Pietro Berkes · Maneesh Sahani -
2007 Poster: Modeling Natural Sounds with Modulation Cascade Processes »
Richard Turner · Maneesh Sahani -
2007 Poster: On Sparsity and Overcompleteness in Image Models »
Pietro Berkes · Richard Turner · Maneesh Sahani