Explainable Time-Series Prediction Using a Residual Network and Gradient-Based Methods
نویسندگان
چکیده
Researchers are employing deep learning (DL) in many fields, and the scope of its application is expanding. However, because understanding rationale validity DL decisions difficult, a model occasionally called black-box model. Here, we focus on DL-based explainable time-series prediction We propose based long short-term memory (LSTM) followed by convolutional neural network (CNN) with residual connection, referred to as LSTM-resCNN. In comparison one-dimensional CNN, bidirectional LSTM, CNN-LSTM, LSTM-CNN, MTEX-CNN models, proposed LSTM-resCNN performs best three datasets fine dust (PM2.5), bike-sharing, bitcoin. Additionally, tested Grad-CAM, Integrated Gradients, gradient-based approaches for explainability. These techniques combined very well Variables time lags that considerably influence can be identified visualized using gradients integrated gradients.
منابع مشابه
Vehicle's velocity time series prediction using neural network
This paper presents the prediction of vehicle's velocity time series using neural networks. For this purpose, driving data is firstly collected in real world traffic conditions in the city of Tehran using advance vehicle location devices installed on private cars. A multi-layer perceptron network is then designed for driving time series forecasting. In addition, the results of this study are co...
متن کاملAvailability Prediction of the Repairable Equipment using Artificial Neural Network and Time Series Models
In this paper, one of the most important criterion in public services quality named availability is evaluated by using artificial neural network (ANN). In addition, the availability values are predicted for future periods by using exponential weighted moving average (EWMA) scheme and some time series models (TSM) including autoregressive (AR), moving average (MA) and autoregressive moving avera...
متن کاملvehicle's velocity time series prediction using neural network
this paper presents the prediction of vehicle's velocity time series using neural networks. for this purpose, driving data is firstly collected in real world traffic conditions in the city of tehran using advance vehicle location devices installed on private cars. a multi-layer perceptron network is then designed for driving time series forecasting. in addition, the results of this study a...
متن کاملSome New Methods for Prediction of Time Series by Wavelets
Extended Abstract. Forecasting is one of the most important purposes of time series analysis. For many years, classical methods were used for this aim. But these methods do not give good performance results for real time series due to non-linearity and non-stationarity of these data sets. On one hand, most of real world time series data display a time-varying second order structure. On th...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Access
سال: 2022
ISSN: ['2169-3536']
DOI: https://doi.org/10.1109/access.2022.3213926