Open AccessProceedings Article
Dropout as a Bayesian approximation: representing model uncertainty in deep learning
Yarin Gal,Zoubin Ghahramani +1 more
- pp 1050-1059
Reads0
Chats0
TLDR
A new theoretical framework is developed casting dropout training in deep neural networks (NNs) as approximate Bayesian inference in deep Gaussian processes, which mitigates the problem of representing uncertainty in deep learning without sacrificing either computational complexity or test accuracy.Abstract:
Deep learning tools have gained tremendous attention in applied machine learning. However such tools for regression and classification do not capture model uncertainty. In comparison, Bayesian models offer a mathematically grounded framework to reason about model uncertainty, but usually come with a prohibitive computational cost. In this paper we develop a new theoretical framework casting dropout training in deep neural networks (NNs) as approximate Bayesian inference in deep Gaussian processes. A direct result of this theory gives us tools to model uncertainty with dropout NNs - extracting information from existing models that has been thrown away so far. This mitigates the problem of representing uncertainty in deep learning without sacrificing either computational complexity or test accuracy. We perform an extensive study of the properties of dropout's uncertainty. Various network architectures and nonlinearities are assessed on tasks of regression and classification, using MNIST as an example. We show a considerable improvement in predictive log-likelihood and RMSE compared to existing state-of-the-art methods, and finish by using dropout's uncertainty in deep reinforcement learning.read more
Citations
More filters
Journal ArticleDOI
AI Perspectives in Smart Cities and Communities to Enable Road Vehicle Automation and Smart Traffic Control
Cristofer Englund,Eren Erdal Aksoy,Fernando Alonso-Fernandez,Martin Cooney,Sepideh Pashami,Björn Åstrand +5 more
TL;DR: Perception, smart traffic control and driver modeling are described along with open research challenges and standardization to help introduce advanced driver assistance systems and automated vehicle functionality in traffic.
Journal ArticleDOI
Multi-View Representation Learning With Deep Gaussian Processes
TL;DR: Experimental results on real-world multi-view data sets verify the effectiveness of the proposed algorithm, which indicates that MvDGPs can integrate the complementary information in multiple views to discover a good representation of the data.
Journal ArticleDOI
Attribute driven inverse materials design using deep learning Bayesian framework
Piyush Tagade,Shashishekar P. Adiga,Shanthi Pandian,Min Sik Park,Krishnan S. Hariharan,Subramanya Mayya Kolake +5 more
TL;DR: A deep learning inverse prediction framework, Structure Learning for Attribute-driven Materials Design Using Novel Conditional Sampling (SLAMDUNCS), for efficient and accurate prediction of molecules exhibiting target properties, which represents an important step in deep learning based completely automated materials discovery.
Book ChapterDOI
Self-Loop Uncertainty: A Novel Pseudo-Label for Semi-supervised Medical Image Segmentation.
TL;DR: Li et al. as mentioned in this paper proposed a semi-supervised approach to train neural networks with limited labeled data and a large quantity of unlabeled images for medical image segmentation, where a novel pseudo-label (namely self-loop uncertainty), generated by recurrently optimizing the neural network with a selfsupervised task, is adopted as the ground-truth for the unlabelled images to augment the training set and boost the segmentation accuracy.
Journal ArticleDOI
Active learning of deep surrogates for PDEs: Application to metasurface design
TL;DR: An active-learning algorithm is presented that reduces the number of simulations required by more than an order of magnitude for an NN surrogate model of optical-surface components compared to uniform random samples and demonstrates how this can be exploited to accelerate large-scale engineering optimization.
References
More filters
Proceedings Article
Adam: A Method for Stochastic Optimization
Diederik P. Kingma,Jimmy Ba +1 more
TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Journal ArticleDOI
Gradient-based learning applied to document recognition
Yann LeCun,Léon Bottou,Léon Bottou,Yoshua Bengio,Yoshua Bengio,Yoshua Bengio,Patrick Haffner +6 more
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Journal Article
Dropout: a simple way to prevent neural networks from overfitting
TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
Journal ArticleDOI
Human-level control through deep reinforcement learning
Volodymyr Mnih,Koray Kavukcuoglu,David Silver,Andrei Rusu,Joel Veness,Marc G. Bellemare,Alex Graves,Martin Riedmiller,Andreas K. Fidjeland,Georg Ostrovski,Stig Petersen,Charles Beattie,Amir Sadik,Ioannis Antonoglou,Helen King,Dharshan Kumaran,Daan Wierstra,Shane Legg,Demis Hassabis +18 more
TL;DR: This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.
Proceedings Article
Auto-Encoding Variational Bayes
Diederik P. Kingma,Max Welling +1 more
TL;DR: A stochastic variational inference and learning algorithm that scales to large datasets and, under some mild differentiability conditions, even works in the intractable case is introduced.