scholarly journals Benchmarking Attention-Based Interpretability of Deep Learning in Multivariate Time Series Predictions

Entropy ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. 143
Author(s):  
Domjan Barić ◽  
Petar Fumić ◽  
Davor Horvatić ◽  
Tomislav Lipic
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Deep Learning ◽  
Deep Neural Networks ◽  
Prediction Performance ◽  
Good Prediction ◽  
Multiple Time ◽  
Multiple Time Series ◽  
Learning Models ◽  
Performance Score

The adaptation of deep learning models within safety-critical systems cannot rely only on good prediction performance but needs to provide interpretable and robust explanations for their decisions. When modeling complex sequences, attention mechanisms are regarded as the established approach to support deep neural networks with intrinsic interpretability. This paper focuses on the emerging trend of specifically designing diagnostic datasets for understanding the inner workings of attention mechanism based deep learning models for multivariate forecasting tasks. We design a novel benchmark of synthetically designed datasets with the transparent underlying generating process of multiple time series interactions with increasing complexity. The benchmark enables empirical evaluation of the performance of attention based deep neural networks in three different aspects: (i) prediction performance score, (ii) interpretability correctness, (iii) sensitivity analysis. Our analysis shows that although most models have satisfying and stable prediction performance results, they often fail to give correct interpretability. The only model with both a satisfying performance score and correct interpretability is IMV-LSTM, capturing both autocorrelations and crosscorrelations between multiple time series. Interestingly, while evaluating IMV-LSTM on simulated data from statistical and mechanistic models, the correctness of interpretability increases with more complex datasets.

scholarly journals Application of Deep Learning Methods to Forecasting Changes in Short-Term Currency Trends

2020 ◽  
Vol 7 (2) ◽  
pp. 75-86
Author(s):  
Vasily D. Derbentsev ◽  
Vitalii S. Bezkorovainyi ◽  
Iryna V. Luniak
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Deep Learning ◽  
Deep Neural Networks ◽  
Short Term Memory ◽  
Learning Models ◽  
Short Term ◽  
Term Memory ◽  
Long Short Term Memory ◽  
Better Than

This study investigates the issues of forecasting changes in short-term currency trends using deep learning models, which is relevant for both the scientific community and for traders and investors. The purpose of this study is to build a model for forecasting the direction of change in the prices of currency quotes based on deep neural networks. The developed architecture was based on the model of valve recurrent node, which is a modification of the model of “Long Short-Term Memory”, but is simpler in terms of the number of parameters and learning time. The forecast calculations of the dynamics of quotations of the currency pair euro/dollar and the most capitalised cryptocurrency Bitcoin/dollar were performed using daily, four-hour and hourly datasets. The obtained results of binary classification (forecast of the direction of trend change) when applying daily and hourly quotations turned out to be generally better than those of time series models or models of neural networks of other architecture (in particular, multilayer perceptron or “Long Short-Term Memory” models). According to the study results, the highest accuracy of classification was for the model of daily quotations for both euro/dollar – about 72%, and for Bitcoin/ dollar – about 69%. For four-hour and hourly time series, the accuracy of classification decreased, which can be explained both by the increase in the impact of “market noise” and the probable overfitting. Computer simulation has demonstrated that models predict a rising trend better than a declining one. The study confirmed the prospects for the application of deep learning models for short-term forecasting of time series of currency quotes. The use of the developed models proved to be effective for both fiat and cryptocurrencies. The proposed system of models based on deep neural networks can be used as a basis for developing an automated trading system in the foreign exchange market

Evolving deep neural networks for Time Series Forecasting

2021 ◽  
Vol 18 (2) ◽  
pp. 40-55
Author(s):  
Lídio Mauro Lima Campos ◽  
◽  
Jherson Haryson Almeida Pereira ◽  
Danilo Souza Duarte ◽  
Roberto Célio Limão Oliveira ◽  
...  
Keyword(s):  
Genetic Algorithm ◽  
Neural Networks ◽  
Time Series ◽  
Deep Neural Networks ◽  
Good Prediction ◽  
Biologically Inspired ◽  
Price Prediction ◽  
Error Measures ◽  
Good Ability ◽  
Proposed Model

The aim of this paper is to introduce a biologically inspired approach that can automatically generate Deep Neural networks with good prediction capacity, smaller error and large tolerance to noises. In order to do this, three biological paradigms were used: Genetic Algorithm (GA), Lindenmayer System and Neural Networks (DNNs). The final sections of the paper present some experiments aimed at investigating the possibilities of the method in the forecast the price of energy in the Brazilian market. The proposed model considers a multi-step ahead price prediction (12, 24, and 36 weeks ahead). The results for MLP and LSTM networks show a good ability to predict peaks and satisfactory accuracy according to error measures comparing with other methods.

A Review of Recent Deep Learning Models in COVID-19 Diagnosis

2021 ◽  
Vol 6 (5) ◽  
pp. 10-15
Author(s):  
Ela Bhattacharya ◽  
D. Bhattacharya
Keyword(s):  
Artificial Intelligence ◽  
Neural Networks ◽  
Deep Learning ◽  
Deep Neural Networks ◽  
Test Results ◽  
Learning Models ◽  
Future Directions ◽  
Human Contact ◽  
The World ◽  
Short Span

COVID-19 has emerged as the latest worrisome pandemic, which is reported to have its outbreak in Wuhan, China. The infection spreads by means of human contact, as a result, it has caused massive infections across 200 countries around the world. Artificial intelligence has likewise contributed to managing the COVID-19 pandemic in various aspects within a short span of time. Deep Neural Networks that are explored in this paper have contributed to the detection of COVID-19 from imaging sources. The datasets, pre-processing, segmentation, feature extraction, classification and test results which can be useful for discovering future directions in the domain of automatic diagnosis of the disease, utilizing artificial intelligence-based frameworks, have been investigated in this paper.

Identification of Thoracic Diseases by Exploiting Deep Neural Networks (Preprint)

2020 ◽  
Author(s):  
Albahli Saleh ◽  
Ali Alkhalifah
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Deep Neural Networks ◽  
Medical Image Analysis ◽  
Medical Community ◽  
Learning Models ◽  
X Ray ◽  
Chest Disease

BACKGROUND To diagnose cardiothoracic diseases, a chest x-ray (CXR) is examined by a radiologist. As more people get affected, doctors are becoming scarce especially in developing countries. However, with the advent of image processing tools, the task of diagnosing these cardiothoracic diseases has seen great progress. A lot of researchers have put in work to see how the problems associated with medical images can be mitigated by using neural networks. OBJECTIVE Previous works used state-of-the-art techniques and got effective results with one or two cardiothoracic diseases but could lead to misclassification. In our work, we adopted GANs to synthesize the chest radiograph (CXR) to augment the training set on multiple cardiothoracic diseases to efficiently diagnose the chest diseases in different classes as shown in Figure 1. In this regard, our major contributions are classifying various cardiothoracic diseases to detect a specific chest disease based on CXR, use the advantage of GANs to overcome the shortages of small training datasets, address the problem of imbalanced data; and implementing optimal deep neural network architecture with different hyper-parameters to improve the model with the best accuracy. METHODS For this research, we are not building a model from scratch due to computational restraints as they require very high-end computers. Rather, we use a Convolutional Neural Network (CNN) as a class of deep neural networks to propose a generative adversarial network (GAN) -based model to generate synthetic data for training the data as the amount of the data is limited. We will use pre-trained models which are models that were trained on a large benchmark dataset to solve a problem similar to the one we want to solve. For example, the ResNet-152 model we used was initially trained on the ImageNet dataset. RESULTS After successful training and validation of the models we developed, ResNet-152 with image augmentation proved to be the best model for the automatic detection of cardiothoracic disease. However, one of the main problems associated with radiographic deep learning projects and research is the scarcity and unavailability of enough datasets which is a key component of all deep learning models as they require a lot of data for training. This is the reason why some of our models had image augmentation to increase the number of images without duplication. As more data are collected in the field of chest radiology, the models could be retrained to improve the accuracies of the models as deep learning models improve with more data. CONCLUSIONS This research employs the advantages of computer vision and medical image analysis to develop an automated model that has the clinical potential for early detection of the disease. Using deep learning models, the research aims to evaluate the effectiveness and accuracy of different convolutional neural network models in the automatic diagnosis of cardiothoracic diseases from x-ray images compared to diagnosis by experts in the medical community.

Network attributes describe a similarity between deep neural networks and large scale brain networks

2019 ◽  
Author(s):  
Kosuke Takagi
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Human Brain ◽  
Large Scale ◽  
Deep Neural Networks ◽  
Distribution Model ◽  
Common Mechanism ◽  
Learning Models ◽  
Connection Weight ◽  
Wide Range

Abstract Despite the recent success of deep learning models in solving various problems, their ability is still limited compared with human intelligence, which has the flexibility to adapt to a changing environment. To obtain a model which achieves adaptability to a wide range of problems and tasks is a challenging problem. To achieve this, an issue that must be addressed is identification of the similarities and differences between the human brain and deep neural networks. In this article, inspired by the human flexibility which might suggest the existence of a common mechanism allowing solution of different kinds of tasks, we consider a general learning process in neural networks, on which no specific conditions and constraints are imposed. Subsequently, we theoretically show that, according to the learning progress, the network structure converges to the state, which is characterized by a unique distribution model with respect to network quantities such as the connection weight and node strength. Noting that the empirical data indicate that this state emerges in the large scale network in the human brain, we show that the same state can be reproduced in a simple example of deep learning models. Although further research is needed, our findings provide an insight into the common inherent mechanism underlying the human brain and deep learning. Thus, our findings provide suggestions for designing efficient learning algorithms for solving a wide variety of tasks in the future.

scholarly journals Integrating Deep Neural Networks and Symbolic Inference for Organic Reactivity Prediction

2020 ◽  
Author(s):  
Wesley Wei Qian ◽  
Nathan T. Russell ◽  
Claire L. W. Simons ◽  
Yunan Luo ◽  
Martin D. Burke ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Probability Distribution ◽  
Deep Neural Networks ◽  
Machine Intelligence ◽  
Data Availability ◽  
Learning Models ◽  
Expert Performance ◽  
Covalent Bonds ◽  
Chemical Reasoning

<div>Accurate <i>in silico</i> models for the prediction of novel chemical reaction outcomes can be used to guide the rapid discovery of new reactivity and enable novel synthesis strategies for newly discovered lead compounds. Recent advances in machine learning, driven by deep learning models and data availability, have shown utility throughout synthetic organic chemistry as a data-driven method for reaction prediction. Here we present a machine-intelligence approach to predict the products of an organic reaction by integrating deep neural networks with a probabilistic and symbolic inference that flexibly enforces chemical constraints and accounts for prior chemical knowledge. We first train a graph convolutional neural network to estimate the likelihood of changes in covalent bonds, hydrogen counts, and formal charges. These estimated likelihoods govern a probability distribution over potential products. Integer Linear Programming is then used to infer the most probable products from the probability distribution subject to heuristic rules such as the octet rule and chemical constraints that reflect a user's prior knowledge. Our approach outperforms previous graph-based neural networks by predicting products with more than 90% accuracy, demonstrates intuitive chemical reasoning through a learned attention mechanism, and provides generalizability across various reaction types. Furthermore, we demonstrate the potential for even higher model accuracy when complemented by expert chemists contributing to the system, boosting both machine and expert performance. The results show the advantages of empowering deep learning models with chemical intuition and knowledge to expedite the drug discovery process.</div>

scholarly journals Exploring deep neural networks via layer-peeled model: Minority collapse in imbalanced training

2021 ◽  
Vol 118 (43) ◽  
pp. e2103091118
Author(s):  
Cong Fang ◽  
Hangfeng He ◽  
Qi Long ◽  
Weijie J. Su
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Deep Neural Networks ◽  
Model Minority ◽  
Tight Frame ◽  
Learning Models ◽  
The Neural Network ◽  
Long Time ◽  
Topmost Layer

In this paper, we introduce the Layer-Peeled Model, a nonconvex, yet analytically tractable, optimization program, in a quest to better understand deep neural networks that are trained for a sufficiently long time. As the name suggests, this model is derived by isolating the topmost layer from the remainder of the neural network, followed by imposing certain constraints separately on the two parts of the network. We demonstrate that the Layer-Peeled Model, albeit simple, inherits many characteristics of well-trained neural networks, thereby offering an effective tool for explaining and predicting common empirical patterns of deep-learning training. First, when working on class-balanced datasets, we prove that any solution to this model forms a simplex equiangular tight frame, which, in part, explains the recently discovered phenomenon of neural collapse [V. Papyan, X. Y. Han, D. L. Donoho, Proc. Natl. Acad. Sci. U.S.A. 117, 24652–24663 (2020)]. More importantly, when moving to the imbalanced case, our analysis of the Layer-Peeled Model reveals a hitherto-unknown phenomenon that we term Minority Collapse, which fundamentally limits the performance of deep-learning models on the minority classes. In addition, we use the Layer-Peeled Model to gain insights into how to mitigate Minority Collapse. Interestingly, this phenomenon is first predicted by the Layer-Peeled Model before being confirmed by our computational experiments.

Sign in / Sign up

Export Citation Format

Share Document