Model confidence or uncertainty is critical in autonomous systems as they directly tie to the safety and trustworthiness of the system. The quantification of uncertainty in the output decisions of deep neural networks (DNNs) is a challenging problem. The Bayesian framework enables the estimation of the predictive uncertainty by introducing probability distributions over the (unknown) network weights; however, the propagation of these high-dimensional distributions through multiple layers and non-linear transformations is mathematically intractable. In this work, we propose an extended variational inference (eVI) framework for convolutional neural network (CNN) based on tensor Normal distributions (TNDs) defined over convolutional kernels. Our proposed eVI framework propagates the first two moments (mean and covariance) of these TNDs through all layers of the CNN. We employ first-order Taylor series linearization to approximate the mean and covariances passing through the non-linear activations. The uncertainty in the output decision is given by the propagated covariance of the predictive distribution. Furthermore, we show, through extensive simulations on the MNIST and CIFAR-10 datasets, that the CNN becomes more robust to Gaussian noise and adversarial attacks.
more »
« less
This content will become publicly available on January 1, 2024
Sinh-arcsinh-normal distributions to add uncertainty to neural network regression tasks: Applications to tropical cyclone intensity forecasts
Abstract A simple method for adding uncertainty to neural network regression tasks in earth science via estimation of a general probability distribution is described. Specifically, we highlight the sinh-arcsinh-normal distributions as particularly well suited for neural network uncertainty estimation. The methodology supports estimation of heteroscedastic, asymmetric uncertainties by a simple modification of the network output and loss function. Method performance is demonstrated by predicting tropical cyclone intensity forecast uncertainty and by comparing two other common methods for neural network uncertainty quantification (i.e., Bayesian neural networks and Monte Carlo dropout). The simple approach described here is intuitive and applicable when no prior exists and one just wishes to parameterize the output and its uncertainty according to some previously defined family of distributions. The authors believe it will become a powerful, go-to method moving forward.
more »
« less
- Award ID(s):
- 2019758
- NSF-PAR ID:
- 10422653
- Date Published:
- Journal Name:
- Environmental Data Science
- Volume:
- 2
- ISSN:
- 2634-4602
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Lai, Yuan (Ed.)Mistrust is a major barrier to implementing deep learning in healthcare settings. Entrustment could be earned by conveying model certainty, or the probability that a given model output is accurate, but the use of uncertainty estimation for deep learning entrustment is largely unexplored, and there is no consensus regarding optimal methods for quantifying uncertainty. Our purpose is to critically evaluate methods for quantifying uncertainty in deep learning for healthcare applications and propose a conceptual framework for specifying certainty of deep learning predictions. We searched Embase, MEDLINE, and PubMed databases for articles relevant to study objectives, complying with PRISMA guidelines, rated study quality using validated tools, and extracted data according to modified CHARMS criteria. Among 30 included studies, 24 described medical imaging applications. All imaging model architectures used convolutional neural networks or a variation thereof. The predominant method for quantifying uncertainty was Monte Carlo dropout, producing predictions from multiple networks for which different neurons have dropped out and measuring variance across the distribution of resulting predictions. Conformal prediction offered similar strong performance in estimating uncertainty, along with ease of interpretation and application not only to deep learning but also to other machine learning approaches. Among the six articles describing non-imaging applications, model architectures and uncertainty estimation methods were heterogeneous, but predictive performance was generally strong, and uncertainty estimation was effective in comparing modeling methods. Overall, the use of model learning curves to quantify epistemic uncertainty (attributable to model parameters) was sparse. Heterogeneity in reporting methods precluded the performance of a meta-analysis. Uncertainty estimation methods have the potential to identify rare but important misclassifications made by deep learning models and compare modeling methods, which could build patient and clinician trust in deep learning applications in healthcare. Efficient maturation of this field will require standardized guidelines for reporting performance and uncertainty metrics.more » « less
-
Table of Contents: Foreword by the CI 2016 Workshop Chairs …………………………………vi Foreword by the CI 2016 Steering Committee ..…………………………..…..viii List of Organizing Committee ………………………….……....x List of Registered Participants .………………………….……..xi Acknowledgement of Sponsors ……………………………..…xiv Hackathon and Workshop Agenda .………………………………..xv Hackathon Summary .………………………….…..xviii Invited talks - abstracts and links to presentations ………………………………..xxi Proceedings: 34 short research papers ……………………………….. 1-135 Papers 1. BAYESIAN MODELS FOR CLIMATE RECONSTRUCTION FROM POLLEN RECORDS ..................................... 1 Lasse Holmström, Liisa Ilvonen, Heikki Seppä, Siim Veski 2. ON INFORMATION CRITERIA FOR DYNAMIC SPATIO-TEMPORAL CLUSTERING ..................................... 5 Ethan D. Schaeffer, Jeremy M. Testa, Yulia R. Gel, Vyacheslav Lyubchich 3. DETECTING MULTIVARIATE BIOSPHERE EXTREMES ..................................... 9 Yanira Guanche García, Erik Rodner, Milan Flach, Sebastian Sippel, Miguel Mahecha, Joachim Denzler 4. SPATIO-TEMPORAL GENERATIVE MODELS FOR RAINFALL OVER INDIA ..................................... 13 Adway Mitra 5. A NONPARAMETRIC COPULA BASED BIAS CORRECTION METHOD FOR STATISTICAL DOWNSCALING ..................................... 17 Yi Li, Adam Ding, Jennifer Dy 6. DETECTING AND PREDICTING BEAUTIFUL SUNSETS USING SOCIAL MEDIA DATA ..................................... 21 Emma Pierson 7. OCEANTEA: EXPLORING OCEAN-DERIVED CLIMATE DATA USING MICROSERVICES ..................................... 25 Arne N. Johanson, Sascha Flögel, Wolf-Christian Dullo, Wilhelm Hasselbring 8. IMPROVED ANALYSIS OF EARTH SYSTEM MODELS AND OBSERVATIONS USING SIMPLE CLIMATE MODELS ..................................... 29 Balu Nadiga, Nathan Urban 9. SYNERGY AND ANALOGY BETWEEN 15 YEARS OF MICROWAVE SST AND ALONG-TRACK SSH ..................................... 33 Pierre Tandeo, Aitor Atencia, Cristina Gonzalez-Haro 10. PREDICTING EXECUTION TIME OF CLIMATE-DRIVEN ECOLOGICAL FORECASTING MODELS ..................................... 37 Scott Farley and John W. Williams 11. SPATIOTEMPORAL ANALYSIS OF SEASONAL PRECIPITATION OVER US USING CO-CLUSTERING ..................................... 41 Mohammad Gorji–Sefidmazgi, Clayton T. Morrison 12. PREDICTION OF EXTREME RAINFALL USING HYBRID CONVOLUTIONAL-LONG SHORT TERM MEMORY NETWORKS ..................................... 45 Sulagna Gope, Sudeshna Sarkar, Pabitra Mitra 13. SPATIOTEMPORAL PATTERN EXTRACTION WITH DATA-DRIVEN KOOPMAN OPERATORS FOR CONVECTIVELY COUPLED EQUATORIAL WAVES ..................................... 49 Joanna Slawinska, Dimitrios Giannakis 14. COVARIANCE STRUCTURE ANALYSIS OF CLIMATE MODEL OUTPUT ..................................... 53 Chintan Dalal, Doug Nychka, Claudia Tebaldi 15. SIMPLE AND EFFICIENT TENSOR REGRESSION FOR SPATIOTEMPORAL FORECASTING ..................................... 57 Rose Yu, Yan Liu 16. TRACKING OF TROPICAL INTRASEASONAL CONVECTIVE ANOMALIES ..................................... 61 Bohar Singh, James L. Kinter 17. ANALYSIS OF AMAZON DROUGHTS USING SUPERVISED KERNEL PRINCIPAL COMPONENT ANALYSIS ..................................... 65 Carlos H. R. Lima, Amir AghaKouchak 18. A BAYESIAN PREDICTIVE ANALYSIS OF DAILY PRECIPITATION DATA ..................................... 69 Sai K. Popuri, Nagaraj K. Neerchal, Amita Mehta 19. INCORPORATING PRIOR KNOWLEDGE IN SPATIO-TEMPORAL NEURAL NETWORK FOR CLIMATIC DATA ..................................... 73 Arthur Pajot, Ali Ziat, Ludovic Denoyer, Patrick Gallinari 20. DIMENSIONALITY-REDUCTION OF CLIMATE DATA USING DEEP AUTOENCODERS ..................................... 77 Juan A. Saenz, Nicholas Lubbers, Nathan M. Urban 21. MAPPING PLANTATION IN INDONESIA ..................................... 81 Xiaowei Jia, Ankush Khandelwal, James Gerber, Kimberly Carlson, Paul West, Vipin Kumar 22. FROM CLIMATE DATA TO A WEIGHTED NETWORK BETWEEN FUNCTIONAL DOMAINS ..................................... 85 Ilias Fountalis, Annalisa Bracco, Bistra Dilkina, Constantine Dovrolis 23. EMPLOYING SOFTWARE ENGINEERING PRINCIPLES TO ENHANCE MANAGEMENT OF CLIMATOLOGICAL DATASETS FOR CORAL REEF ANALYSIS ..................................... 89 Mark Jenne, M.M. Dalkilic, Claudia Johnson 24. Profiler Guided Manual Optimization for Accelerating Cholesky Decomposition on R Environment ..................................... 93 V.B. Ramakrishnaiah, R.P. Kumar, J. Paige, D. Hammerling, D. Nychka 25. GLOBAL MONITORING OF SURFACE WATER EXTENT DYNAMICS USING SATELLITE DATA ..................................... 97 Anuj Karpatne, Ankush Khandelwal and Vipin Kumar 26. TOWARD QUANTIFYING TROPICAL CYCLONE RISK USING DIAGNOSTIC INDICES .................................... 101 Erica M. Staehling and Ryan E. Truchelut 27. OPTIMAL TROPICAL CYCLONE INTENSITY ESTIMATES WITH UNCERTAINTY FROM BEST TRACK DATA .................................... 105 Suz Tolwinski-Ward 28. EXTREME WEATHER PATTERN DETECTION USING DEEP CONVOLUTIONAL NEURAL NETWORK .................................... 109 Yunjie Liu, Evan Racah, Prabhat, Amir Khosrowshahi, David Lavers, Kenneth Kunkel, Michael Wehner, William Collins 29. INFORMATION TRANSFER ACROSS TEMPORAL SCALES IN ATMOSPHERIC DYNAMICS .................................... 113 Nikola Jajcay and Milan Paluš 30. Identifying precipitation regimes in China using model-based clustering of spatial functional data .................................... 117 Haozhe Zhang, Zhengyuan Zhu, Shuiqing Yin 31. RELATIONAL RECURRENT NEURAL NETWORKS FOR SPATIOTEMPORAL INTERPOLATION FROM MULTI-RESOLUTION CLIMATE DATA .................................... 121 Guangyu Li, Yan Liu 32. OBJECTIVE SELECTION OF ENSEMBLE BOUNDARY CONDITIONS FOR CLIMATE DOWNSCALING .................................... 124 Andrew Rhines, Naomi Goldenson 33. LONG-LEAD PREDICTION OF EXTREME PRECIPITATION CLUSTER VIA A SPATIO-TEMPORAL CONVOLUTIONAL NEURAL NETWORK .................................... 128 Yong Zhuang, Wei Ding 34. MULTIPLE INSTANCE LEARNING FOR BURNED AREA MAPPING USING MULTI –TEMPORAL REFLECTANCE DATA .................................... 132 Guruprasad Nayak, Varun Mithal, Vipin Kumarmore » « less
-
more » « less
Abstract The earth system is exceedingly complex and often chaotic in nature, making prediction incredibly challenging: we cannot expect to make perfect predictions all of the time. Instead, we look for specific states of the system that lead to more predictable behavior than others, often termed “forecasts of opportunity.” When these opportunities are not present, scientists need prediction systems that are capable of saying “I don't know.” We introduce a novel loss function, termed “abstention loss,” that allows neural networks to identify forecasts of opportunity for regression problems. The abstention loss works by incorporating uncertainty in the network's prediction to identify the more confident samples and abstain (say “I don't know”) on the less confident samples. The abstention loss is designed to determine the optimal abstention fraction, or abstain on a user‐defined fraction using a standard adaptive controller. Unlike many methods for attaching uncertainty to neural network predictions post‐training, the abstention loss is applied during training to preferentially learn from the more confident samples. The abstention loss is built upon nonlinear heteroscedastic regression, a standard computer science method. While nonlinear heteroscedastic regression is a simple yet powerful tool for incorporating uncertainty in regression problems, we demonstrate that the abstention loss outperforms it for the synthetic climate use cases explored here. The implementation of the proposed abstention loss is straightforward in most network architectures designed for regression, as it only requires modification of the output layer and loss function.
-
Each year a growing number of wind farms are being added to power grids to generate sustainable energy. The power curve of a wind turbine, which exhibits the relationship between generated power and wind speed, plays a major role in assessing the performance of a wind farm. Neural networks have been used for power curve estimation. However, they do not produce a confidence measure for their output, unless computationally prohibitive Bayesian methods are used. In this paper, a probabilistic neural network with Monte Carlo dropout is considered to quantify the model or epistemic uncertainty of the power curve estimation. This approach offers a minimal increase in computational complexity and thus evaluation time. Furthermore, by adding a probabilistic loss function, the noise or aleatoric uncertainty in the data is estimated. The developed network captures both model and noise uncertainty which are found to be useful tools in assessing performance. Also, the developed network is compared with the existing ones across a public domain dataset showing superior performance in terms of prediction accuracy. The results obtained indicate that the developed network provides the quantification of uncertainty while maintaining accurate power estimation.more » « less
