scholarly journals Auditing and Debugging Deep Learning Models via Flip Points: Individual-Level and Group-Level Analysis

La Matematica ◽  
2021 ◽  
Author(s):  
Roozbeh Yousefzadeh ◽  
Dianne P. O’Leary
Keyword(s):  
Deep Learning ◽  
Synthetic Data ◽  
Numerical Linear Algebra ◽  
Optimization Techniques ◽  
Mathematical Function ◽  
Learning Models ◽  
Individual Level ◽  
Deep Model ◽  
Social Applications ◽  
Decision Boundaries

AbstractDeep learning models have been criticized for their lack of easy interpretation, which undermines confidence in their use for important applications. Nevertheless, they are consistently utilized in many applications, consequential to humans’ lives, usually because of their better performance. Therefore, there is a great need for computational methods that can explain, audit, and debug such models. Here, we use flip points to accomplish these goals for deep learning classifiers used in social applications. A trained deep learning classifier is a mathematical function that maps inputs to classes. By way of training, the function partitions its domain and assigns a class to each of the partitions. Partitions are defined by the decision boundaries which are expected to be geometrically complex. This complexity is usually what makes deep learning models powerful classifiers. Flip points are points on those boundaries and, therefore, the key to understanding and changing the functional behavior of models. We use advanced numerical optimization techniques and state-of-the-art methods in numerical linear algebra, such as rank determination and reduced-order models to compute and analyze them. The resulting insight into the decision boundaries of a deep model can clearly explain the model’s output on the individual level, via an explanation report that is understandable by non-experts. We also develop a procedure to understand and audit model behavior towards groups of people. We show that examining decision boundaries of models in certain subspaces can reveal hidden biases that are not easily detectable. Flip points can also be used as synthetic data to alter the decision boundaries of a model and improve their functional behaviors. We demonstrate our methods by investigating several models trained on standard datasets used in social applications of machine learning. We also identify the features that are most responsible for particular classifications and misclassifications. Finally, we discuss the implications of our auditing procedure in the public policy domain.

scholarly journals A Comparative Analysis of Novel Deep Learning and Ensemble Learning Models to Predict the Allergenicity of Food Proteins

Foods ◽  
2021 ◽  
Vol 10 (4) ◽  
pp. 809
Author(s):  
Liyang Wang ◽  
Dantong Niu ◽  
Xinjie Zhao ◽  
Xiaoya Wang ◽  
Mengzhen Hao ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Ensemble Learning ◽  
Food Allergen ◽  
Gradient Boosting ◽  
Learning Models ◽  
Food Proteins ◽  
Deep Model ◽  
Allergen Identification ◽  
Machine Learning Models

Traditional food allergen identification mainly relies on in vivo and in vitro experiments, which often needs a long period and high cost. The artificial intelligence (AI)-driven rapid food allergen identification method has solved the above mentioned some drawbacks and is becoming an efficient auxiliary tool. Aiming to overcome the limitations of lower accuracy of traditional machine learning models in predicting the allergenicity of food proteins, this work proposed to introduce deep learning model—transformer with self-attention mechanism, ensemble learning models (representative as Light Gradient Boosting Machine (LightGBM) eXtreme Gradient Boosting (XGBoost)) to solve the problem. In order to highlight the superiority of the proposed novel method, the study also selected various commonly used machine learning models as the baseline classifiers. The results of 5-fold cross-validation showed that the area under the receiver operating characteristic curve (AUC) of the deep model was the highest (0.9578), which was better than the ensemble learning and baseline algorithms. But the deep model need to be pre-trained, and the training time is the longest. By comparing the characteristics of the transformer model and boosting models, it can be analyzed that, each model has its own advantage, which provides novel clues and inspiration for the rapid prediction of food allergens in the future.

scholarly journals An Interpretation Architecture for Deep Learning Models with the Application of COVID-19 Diagnosis

Entropy ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. 204
Author(s):  
Yuchai Wan ◽  
Hongen Zhou ◽  
Xun Zhang
Keyword(s):  
Deep Learning ◽  
Predictive Value ◽  
Classification Performance ◽  
Learning Approaches ◽  
Learning Models ◽  
Visual Interpretation ◽  
Deep Model ◽  
Diagnostic Decision ◽  
Sensitivity Specificity ◽  
Learned Features

The Coronavirus disease 2019 (COVID-19) has become one of the threats to the world. Computed tomography (CT) is an informative tool for the diagnosis of COVID-19 patients. Many deep learning approaches on CT images have been proposed and brought promising performance. However, due to the high complexity and non-transparency of deep models, the explanation of the diagnosis process is challenging, making it hard to evaluate whether such approaches are reliable. In this paper, we propose a visual interpretation architecture for the explanation of the deep learning models and apply the architecture in COVID-19 diagnosis. Our architecture designs a comprehensive interpretation about the deep model from different perspectives, including the training trends, diagnostic performance, learned features, feature extractors, the hidden layers, the support regions for diagnostic decision, and etc. With the interpretation architecture, researchers can make a comparison and explanation about the classification performance, gain insight into what the deep model learned from images, and obtain the supports for diagnostic decisions. Our deep model achieves the diagnostic result of 94.75%, 93.22%, 96.69%, 97.27%, and 91.88% in the criteria of accuracy, sensitivity, specificity, positive predictive value, and negative predictive value, which are 8.30%, 4.32%, 13.33%, 10.25%, and 6.19% higher than that of the compared traditional methods. The visualized features in 2-D and 3-D spaces provide the reasons for the superiority of our deep model. Our interpretation architecture would allow researchers to understand more about how and why deep models work, and can be used as interpretation solutions for any deep learning models based on convolutional neural network. It can also help deep learning methods to take a step forward in the clinical COVID-19 diagnosis field.

Addressing model uncertainty in probabilistic forecasting using Monte Carlo dropout

2020 ◽  
Vol 24 ◽  
pp. 185-205
Author(s):  
Cristián Serpell ◽  
Ignacio A. Araya ◽  
Carlos Valle ◽  
Héctor Allende
Keyword(s):  
Monte Carlo ◽  
Deep Learning ◽  
Model Uncertainty ◽  
Variational Approach ◽  
Variance Estimation ◽  
Learning Models ◽  
Probabilistic Forecasting ◽  
Deep Model ◽  
Uncertain Future ◽  
Mean Variance

In recent years, deep learning models have been developed to address probabilistic forecasting tasks, assuming an implicit stochastic process that relates past observed values to uncertain future values. These models are capable of capturing the inherent uncertainty of the underlying process, but they ignore the model uncertainty that comes from the fact of not having infinite data. This work proposes addressing the model uncertainty problem using Monte Carlo dropout, a variational approach that assigns distributions to the weights of a neural network instead of simply using fixed values. This allows to easily adapt common deep learning models currently in use to produce better probabilistic forecasting estimates, in terms of their consideration of uncertainty. The proposal is validated for prediction intervals estimation on seven energy time series, using a popular probabilistic model called Mean Variance Estimation (MVE), as the deep model adapted using the technique.

Levenshtein Augmentation Improves Performance of SMILES Based Deep-Learning Synthesis Prediction

2020 ◽  
Author(s):  
Dean Sumner ◽  
Jiazhen He ◽  
Amol Thakkar ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum
Keyword(s):  
Neural Networks ◽  
Pattern Recognition ◽  
Deep Learning ◽  
Recurrent Neural Networks ◽  
Data Augmentation ◽  
State Of The Art ◽  
Sequence Similarity ◽  
Learning Models ◽  
Underlying Network

<p>SMILES randomization, a form of data augmentation, has previously been shown to increase the performance of deep learning models compared to non-augmented baselines. Here, we propose a novel data augmentation method we call “Levenshtein augmentation” which considers local SMILES sub-sequence similarity between reactants and their respective products when creating training pairs. The performance of Levenshtein augmentation was tested using two state of the art models - transformer and sequence-to-sequence based recurrent neural networks with attention. Levenshtein augmentation demonstrated an increase performance over non-augmented, and conventionally SMILES randomization augmented data when used for training of baseline models. Furthermore, Levenshtein augmentation seemingly results in what we define as <i>attentional gain </i>– an enhancement in the pattern recognition capabilities of the underlying network to molecular motifs.</p>

Improving the Accuracy of Protein-Ligand Binding Affinity Prediction by Deep Learning Models: Benchmark and Model

2019 ◽  
Author(s):  
Mohammad Rezaei ◽  
Yanjun Li ◽  
Xiaolin Li ◽  
Chenglong Li
Keyword(s):  
Deep Learning ◽  
Drug Design ◽  
Binding Affinity ◽  
Benchmark Dataset ◽  
Rational Drug Design ◽  
Learning Models ◽  
Structure Based Drug Design ◽  
Binding Affinity Prediction ◽  
Affinity Prediction ◽  
Rational Drug

<b>Introduction:</b> The ability to discriminate among ligands binding to the same protein target in terms of their relative binding affinity lies at the heart of structure-based drug design. Any improvement in the accuracy and reliability of binding affinity prediction methods decreases the discrepancy between experimental and computational results.<br><b>Objectives:</b> The primary objectives were to find the most relevant features affecting binding affinity prediction, least use of manual feature engineering, and improving the reliability of binding affinity prediction using efficient deep learning models by tuning the model hyperparameters.<br><b>Methods:</b> The binding site of target proteins was represented as a grid box around their bound ligand. Both binary and distance-dependent occupancies were examined for how an atom affects its neighbor voxels in this grid. A combination of different features including ANOLEA, ligand elements, and Arpeggio atom types were used to represent the input. An efficient convolutional neural network (CNN) architecture, DeepAtom, was developed, trained and tested on the PDBbind v2016 dataset. Additionally an extended benchmark dataset was compiled to train and evaluate the models.<br><b>Results: </b>The best DeepAtom model showed an improved accuracy in the binding affinity prediction on PDBbind core subset (Pearson’s R=0.83) and is better than the recent state-of-the-art models in this field. In addition when the DeepAtom model was trained on our proposed benchmark dataset, it yields higher correlation compared to the baseline which confirms the value of our model.<br><b>Conclusions:</b> The promising results for the predicted binding affinities is expected to pave the way for embedding deep learning models in virtual screening and rational drug design fields.

Data science in economics: comprehensive review of advanced machine learning and deep learning methods

2020 ◽  
Author(s):  
Saeed Nosratabadi ◽  
Amir Mosavi ◽  
Puhong Duan ◽  
Pedram Ghamisi ◽  
Ferdinand Filip ◽  
...  
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Data Science ◽  
State Of The Art ◽  
Science Methods ◽  
Learning Models ◽  
Diverse Range ◽  
Hybrid Machine ◽  
Economics Research

This paper provides a state-of-the-art investigation of advances in data science in emerging economic applications. The analysis was performed on novel data science methods in four individual classes of deep learning models, hybrid deep learning models, hybrid machine learning, and ensemble models. Application domains include a wide and diverse range of economics research from the stock market, marketing, and e-commerce to corporate banking and cryptocurrency. Prisma method, a systematic literature review methodology, was used to ensure the quality of the survey. The findings reveal that the trends follow the advancement of hybrid models, which, based on the accuracy metric, outperform other learning algorithms. It is further expected that the trends will converge toward the advancements of sophisticated hybrid deep learning models.

A Study on the Auxiliary Diagnosis of Thyroid Disease Images Based on Multiple Dimensional Deep Learning Algorithms

2020 ◽  
Vol 16 (3) ◽  
pp. 199-205
Author(s):  
Yuejun Liu ◽  
Yifei Xu ◽  
Xiangzheng Meng ◽  
Xuguang Wang ◽  
Tianxu Bai
Keyword(s):  
Deep Learning ◽  
Learning Algorithms ◽  
Region Of Interest ◽  
Classification Performance ◽  
Thyroid Diseases ◽  
Great Success ◽  
Learning Models ◽  
Good Classification Performance ◽  
Spect Images

Background: Medical imaging plays an important role in the diagnosis of thyroid diseases. In the field of machine learning, multiple dimensional deep learning algorithms are widely used in image classification and recognition, and have achieved great success. Objective: The method based on multiple dimensional deep learning is employed for the auxiliary diagnosis of thyroid diseases based on SPECT images. The performances of different deep learning models are evaluated and compared. Methods: Thyroid SPECT images are collected with three types, they are hyperthyroidism, normal and hypothyroidism. In the pre-processing, the region of interest of thyroid is segmented and the amount of data sample is expanded. Four CNN models, including CNN, Inception, VGG16 and RNN, are used to evaluate deep learning methods. Results: Deep learning based methods have good classification performance, the accuracy is 92.9%-96.2%, AUC is 97.8%-99.6%. VGG16 model has the best performance, the accuracy is 96.2% and AUC is 99.6%. Especially, the VGG16 model with a changing learning rate works best. Conclusion: The standard CNN, Inception, VGG16, and RNN four deep learning models are efficient for the classification of thyroid diseases with SPECT images. The accuracy of the assisted diagnostic method based on deep learning is higher than that of other methods reported in the literature.

Sign in / Sign up

Export Citation Format

Share Document