Visualization of neural network predictions for weather forecasting
Pages 61 - 68
Abstract
Recurrent neural networks are prime candidates for learning relationships and evolutions in multi-dimensional time series data. The performance of such a network is judged by the loss function, which is aggregated into a single scalar value that decreases during successful training. Observing only this number hides the variation that occurs within the typically large training and testing data sets. Understanding these variations is of highest importance to adjust hyperparameters of the network, such as the number of neurons, number of layers or even to adjust the training set to include more representative examples. In this paper, we design a comprehensive and interactive system that allows to study the output of recurrent neural networks on both the complete training data as well as the testing data. We follow a coarse-to-fine strategy, providing overviews of annual, monthly and daily patterns in the time series and directly support a comparison of different hyperparameter settings. We applied our method to a recurrent convolutional neural network that was trained and tested on 25 years of climate data to forecast meteorological attributes, such as temperature, pressure and wind speed. The presented visualization system helped us to quickly assess, adjust and improve the network design.
References
[1]
{BB12} Bergstra J., Bengio Y.: Random search for hyper-parameter optimization. Journal of Machine Learning Research 13, Feb (2012), 281--305. 2
[2]
{BBM*15} Bach S., Binder A., Montavon G., Klauschen F., Müller K.-R., Samek W.: On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PloS one 10, 7 (2015), e0130140. 1
[3]
{BSF94} Bengio Y., Simard P., Frasconi P.: Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks 5, 2 (Mar 1994), 157--166. 2
[4]
{Cho15} Chollet F.: Keras. https://github.com/fchollet/keras, 2015. access date: 12 June 2017. 5
[5]
{Cyn17} Cynthia A.: Brewer. http://colorbrewer2.org, 2017. access date: 12 June 2017. 4
[6]
{DUS*11} Dee D. P., Uppala S. M., Simmons A. J., Berrisford P., Poli P., Kobayashi S., Andrae, et al.: The ERA-Interim reanalysis: configuration and performance of the data assimilation system. Quarterly Journal of the Royal Meteorological Society 137, 656 (2011), 553--597. 4
[7]
{ETFD97} Edwards T., Tansley D., Frank R., Davey N.: Traffic trends analysis using neural networks. In Procs of the Int. Workshop on Applications of Neural Networks to Telecommunications (1997). 1
[8]
{GGT17} Günther T., Gross M., Theisel H.: Generic objective vortices for flow visualization. ACM Transactions on Graphics (Proc. SIGGRAPH) 36, 4 (2017), 141:1--141:11. 5, 7
[9]
{GKH15} Grover A., Kapoor A., Horvitz E.: A deep hybrid model for weather forecasting. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2015), ACM, pp. 379--386. 1
[10]
{GMH13} Graves A., Mohamed A.-r., Hinton G.: Speech recognition with deep recurrent neural networks. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2013), IEEE, pp. 6645--6649. 2
[11]
{Hau06} Hauser H.: Generalizing focus+context visualization. In Scientific visualization: The visual extraction of knowledge from data. Springer, 2006, pp. 305--327. 3
[12]
{HRLD15} Hossain M., Rekabdar B., Louis S. J., Dascalu S.: Forecasting the weather of Nevada: A deep learning approach. In International Joint Conference on Neural Networks (IJCNN) (2015), IEEE, pp. 1--6. 1
[13]
{HS97} Hochreiter S., Schmidhuber J.: Long short-term memory. Neural Comput. 9, 8 (Nov. 1997), 1735--1780. 2
[14]
{KB96} Kaastra I., Boyd M.: Designing a neural network for forecasting financial and economic time series. Neurocomputing 10, 3 (1996), 215--236. 1
[15]
{KJL15} Karpathy A., Johnson J., Li F.: Visualizing and understanding recurrent networks. CoRR abs/1506.02078 (2015). 1
[16]
{LSL*17} Liu M., Shi J., Li Z., Li C., Zhu J., Liu S.: Towards better analysis of deep convolutional neural networks. IEEE Trans. on Vis. and Computer Graphics (Proc. IEEE VAST 2016) 23, 1 (Jan 2017), 91--100. 1
[17]
{SMH11} Sutskever I., Martens J., Hinton G. E.: Generating text with recurrent neural networks. In Proceedings of the 28th International Conference on Machine Learning (ICML-11) (2011), pp. 1017--1024. 2
[18]
{SVZ14} Simonyan K., Vedaldi A., Zisserman A.: Deep inside convolutional networks: Visualising image classification models and saliency maps. In ICLR Workshop Papers (2014). 1
[19]
{SWG05} Schmidhuber J., Wierstra D., Gomez F.: Evolino: Hybrid neuroevolution/optimal linear search for sequence learning. In Proc. International Joint Conference on Artificial Intelligence (2005), Morgan Kaufmann Publishers Inc., pp. 853--858. 2
[20]
{SWLL13} Sun G.-D., Wu Y.-C., Liang R.-H., Liu S.-X.: A survey of visual analytics techniques and applications: State-of-the-art research and future challenges. Journal of Computer Science and Technology 28, 5 (2013), 852--867. 3
[21]
{TN65} Truesdell C., Noll W.: The nonlinear field theories of mechanics. Handbuch der Physik, Band III/3, e by Flugge, S., (ed.), Springer-Verlag, Berlin, 1965. 5
[22]
{VWVS99} Van Wijk J. J., Van Selow E. R.: Cluster and calendar based visualization of time series data. In Proc. IEEE Symposium on Information Visualization (1999), pp. 4--9. 3
[23]
{Wal17} Walser A.: Personal communication, 2017. MeteoSwiss. 8
[24]
{ZCAW17} Zintgraf L. M., Cohen T. S., Adel T., Welling M.: Visualizing deep neural network decisions: Prediction difference analysis. arXiv preprint arXiv:1702.04595 (2017). 1
- Visualization of neural network predictions for weather forecasting
Recommendations
An ensemble of neural networks for weather forecasting
This study presents the applicability of an ensemble of artificial neural networks (ANNs) and learning paradigms for weather forecasting in southern Saskatchewan, Canada. The proposed ensemble method for weather forecasting has advantages over other ...
Weather Forecasting Using Artificial Neural Network (ANN): A Review
AbstractExtreme weather occurrences provide issues that necessitate the development of technology capable of accurate analysis and exact prediction in order to successfully reduce their effects. Artificial neural networks (ANNs) have appeared in the last ...
Application of Artificial Neural Network for Daily Evaporation Forecasting Using Weather Data
ICAAI '20: Proceedings of the 4th International Conference on Advances in Artificial IntelligenceEstimating evaporation is one of the most important works for meteorologists and hydrologists. This paper employed three artificial neural network (ANN) algorithms, which are linear regression (LR), multi-layer perceptron (MLP) and general regression ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
September 2017
175 pages
ISBN:9783038680499
- Conference Chairs:
- Matthias Hullin,
- Reinhard Klein,
- Thomas Schultz,
- Angela Yao
Publisher
Eurographics Association
Goslar, Germany
Publication History
Published: 25 September 2017
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 12 Jan 2025