Deep Neural Networks (DNNs) need to be both efficient and robust for practical uses. Quantization and structure simplification are promising ways to adapt DNNs to mobile devices, and adversarial training is one of the most successful methods to train robust DNNs. In this work, we aim to realize both advantages by applying a convergent relaxation quantization algorithm, i.e., Binary-Relax (BR), to an adversarially trained robust model, i.e. the ResNets Ensemble via Feynman-Kac Formalism (EnResNet). We discover that high-precision quantization, such as ternary (tnn) or 4-bit, produces sparse DNNs. However, this sparsity is unstructured under adversarial training. To solve the problems that adversarial training jeopardizes DNNs’ accuracy on clean images and break the structure of sparsity, we design a trade-off loss function that helps DNNs preserve natural accuracy and improve channel sparsity. With our newly designed trade-off loss function, we achieve both goals with no reduction of resistance under weak attacks and very minor reduction of resistance under strong adversarial attacks. Together with our model and algorithm selections and loss function design, we provide an integrated approach to produce robust DNNs with high efficiency and accuracy. Furthermore, we provide a missing benchmark on robustness of quantized models.
more »
« less
Theory for Deep Learning Regression Ensembles with Application to Raman Spectroscopy Analysis
Regression ensembles consisting of a collection of base regression models are often used to improve the estimation/prediction performance of a single regression model. It has been shown that the individual accuracy of the base models and the ensemble diversity are the two key factors affecting the performance of an ensemble. In this paper, we derive a theory for regression ensembles that illustrates the subtle trade-off between individual accuracy and ensemble diversity from the perspective of statistical correlations. Then, inspired by our derived theory, we further propose a novel loss function and a training algorithm for deep learning regression ensembles. We then demonstrate the advantage of our training approach over standard regression ensemble methods including random forest and gradient boosting regressors with both benchmark regression problems and chemical sensor problems involving analysis of Raman spectroscopy. Our key contribution is that our loss function and training algorithm is able to manage diversity explicitly in an ensemble, rather than merely allowing diversity to occur by happenstance.
more »
« less
- Award ID(s):
- 2021871
- NSF-PAR ID:
- 10341216
- Date Published:
- Journal Name:
- 2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA)
- Page Range / eLocation ID:
- 1049 to 1056
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Learning nonlinear functions from input-output data pairs is one of the most fundamental problems in machine learning. Recent work has formulated the problem of learning a general nonlinear multivariate function of discrete inputs, as a tensor completion problem with smooth latent factors. We build upon this idea and utilize two ensemble learning techniques to enhance its prediction accuracy. Ensemble methods can be divided into two main groups, parallel and sequential. Bagging also known as bootstrap aggregation is a parallel ensemble method where multiple base models are trained in parallel on different subsets of the data that have been chosen randomly with replacement from the original training data. The output of these models is usually combined and a single prediction is computed using averaging. One of the most popular bagging techniques is random forests. Boosting is a sequential ensemble method where a sequence of base models are fit sequentially to modified versions of the data. Popular boosting algorithms include AdaBoost and Gradient Boosting. We develop two approaches based on these ensemble learning techniques for learning multivariate functions using the Canonical Polyadic Decomposition. We showcase the effectiveness of the proposed ensemble models on several regression tasks and report significant improvements compared to the single model.more » « less
-
Solar flare prediction is a central problem in space weather forecasting and has captivated the attention of a wide spectrum of researchers due to recent advances in both remote sensing as well as machine learning and deep learning approaches. The experimental findings based on both machine and deep learning models reveal significant performance improvements for task specific datasets. Along with building models, the practice of deploying such models to production environments under operational settings is a more complex and often time-consuming process which is often not addressed directly in research settings. We present a set of new heuristic approaches to train and deploy an operational solar flare prediction system for ≥M1.0-class flares with two prediction modes: full-disk and active region-based. In full-disk mode, predictions are performed on full-disk line-of-sight magnetograms using deep learning models whereas in active region-based models, predictions are issued for each active region individually using multivariate time series data instances. The outputs from individual active region forecasts and full-disk predictors are combined to a final full-disk prediction result with a meta-model. We utilized an equal weighted average ensemble of two base learners’ flare probabilities as our baseline meta learner and improved the capabilities of our two base learners by training a logistic regression model. The major findings of this study are: 1) We successfully coupled two heterogeneous flare prediction models trained with different datasets and model architecture to predict a full-disk flare probability for next 24 h, 2) Our proposed ensembling model, i.e., logistic regression, improves on the predictive performance of two base learners and the baseline meta learner measured in terms of two widely used metrics True Skill Statistic (TSS) and Heidke Skill Score (HSS), and 3) Our result analysis suggests that the logistic regression-based ensemble (Meta-FP) improves on the full-disk model (base learner) by ∼9% in terms TSS and ∼10% in terms of HSS. Similarly, it improves on the AR-based model (base learner) by ∼17% and ∼20% in terms of TSS and HSS respectively. Finally, when compared to the baseline meta model, it improves on TSS by ∼10% and HSS by ∼15%.more » « less
-
more » « less
It is increasingly common to encounter prediction tasks in the biomedical sciences for which multiple datasets are available for model training. Common approaches such as pooling datasets before model fitting can produce poor out‐of‐study prediction performance when datasets are heterogeneous. Theoretical and applied work has shown
multistudy ensembling to be a viable alternative that leverages the variability across datasets in a manner that promotes model generalizability. Multistudy ensembling uses a two‐stagestacking strategy which fits study‐specific models and estimates ensemble weights separately. This approach ignores, however, the ensemble properties at the model‐fitting stage, potentially resulting in performance losses. Motivated by challenges in the estimation of COVID‐attributable mortality, we proposeoptimal ensemble construction , an approach to multistudy stacking whereby we jointly estimate ensemble weights and parameters associated with study‐specific models. We prove that limiting cases of our approach yield existing methods such as multistudy stacking and pooling datasets before model fitting. We propose an efficient block coordinate descent algorithm to optimize the loss function. We use our method to perform multicountry COVID‐19 baseline mortality prediction. We show that when little data is available for a country before the onset of the pandemic, leveraging data from other countries can substantially improve prediction accuracy. We further compare and characterize the method's performance in data‐driven simulations and other numerical experiments. Our method remains competitive with or outperforms multistudy stacking and other earlier methods in the COVID‐19 data application and in a range of simulation settings. -
null (Ed.)Training deep neural models in the presence of corrupted supervision is challenging as the corrupted data points may significantly impact the generalization performance. To alleviate this problem, we present an efficient robust algorithm that achieves strong guarantees without any assumption on the type of corruption and provides a unified framework for both classification and regression problems. Unlike many existing approaches that quantify the quality of the data points (e.g., based on their individual loss values), and filter them accordingly, the proposed algorithm focuses on controlling the collective impact of data points on the average gradient. Even when a corrupted data point failed to be excluded by our algorithm, the data point will have a very limited impact on the overall loss, as compared with state-of-the-art filtering methods based on loss values. Extensive experiments on multiple benchmark datasets have demonstrated the robustness of our algorithm under different types of corruption.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ICMLA52953.2021.00172
