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
Proceedings ArticleDOI
What's Missing: A Knowledge Gap Guided Approach for Multi-hop Question Answering.
TL;DR: A novel approach is developed that explicitly identifies the knowledge gap between a key span in the provided knowledge and the answer choices and learns to fill this gap by determining the relationship between the span and an answer choice, based on retrieved knowledge targeting this gap.
Journal ArticleDOI
Ordinal convolutional neural networks for predicting RDoC positive valence psychiatric symptom severity scores
Anthony Rios,Ramakanth Kavuluru +1 more
TL;DR: A novel convolutional neural network model designed to handle ordinal regression tasks on psychiatric notes is proposed which leads to excellent performance on the N-GRID shared task and is also amenable to interpretability using existing model-agnostic approaches.
Proceedings ArticleDOI
Pessimistic Reward Models for Off-Policy Learning in Recommendation
Olivier Jeunen,Bart Goethals +1 more
TL;DR: In this article, a general pessimistic reward modeling approach for off-policy learning in recommendation is proposed and validated using Bayesian uncertainty estimates, which can be used to generate a conservative decision rule.
Journal ArticleDOI
Probabilistic faster R-CNN with stochastic region proposing: Towards object detection and recognition in remote sensing imagery
Dewei Yi,Jinya Su,Wen-Hua Chen +2 more
TL;DR: A probabilistic faster R-CNN algorithm with stochastic region proposing, where a Probabilistic region proposal network, an image classification network, and an object detection network are integrated to detect and locate grasshoppers is proposed.
Proceedings ArticleDOI
Scalable Bayesian Learning of Recurrent Neural Networks for Language Modeling
TL;DR: The authors leverages stochastic gradient Markov Chain Monte Carlo (SGMC) to learn weight uncertainty in recurrent neural networks (RNNs), which yields a principled Bayesian learning algorithm, adding gradient noise during training and model averaging when testing.
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.