Machine and deep learning-based algorithms are the emerging approaches in addressing prediction problems in time series. These techniques have been shown to produce more accurate results than conventional regression-based modeling.
It has been reported that artificial Recurrent Neural Networks (RNN) with memory, such as Long Short-Term Memory (LSTM), are superior compared to Autoregressive Integrated Moving Average (ARIMA) with a large margin. The LSTM-based models incorporate additional “gates” for the purpose of memorizing
longer sequences of input data. The major question is that whether the gates incorporated in the LSTM architecture already offers a good prediction and whether additional training of data would be necessary to further improve the prediction.
Bidirectional LSTMs (BiLSTMs) enable additional training by traversing the input data twice (i.e., 1) left-to-right, and 2) right-to-left). The research question of interest is then whether BiLSTM, with additional training capability, outperforms regular
unidirectional LSTM. This paper reports a behavioral analysis and comparison of BiLSTM and LSTM models. The objective is to explore to what extend additional layers of training of data would be beneficial to tune the involved parameters. The
results show that additional training of data and thus BiLSTM-based modeling offers better predictions than regular LSTM-based models. More specifically, it was observed that BiLSTM models provide better predictions compared to ARIMA and
LSTM models. It was also observed that BiLSTM models reach the equilibrium much slower than LSTM-based models.
more »
« less
A Comparison of ARIMA and LSTM in Forecasting Time Series
Forecasting time series data is an important subject in economics, business, and finance. Traditionally, there are several techniques to effectively forecast the next lag of time series data such as univariate Autoregressive (AR), univariate Moving Average (MA), Simple Exponential Smoothing (SES), and more notably Autoregressive Integrated Moving Average (ARIMA) with its many variations. In particular, ARIMA model has demonstrated its outperformance in precision and
accuracy of predicting the next lags of time series. With the recent advancement in computational power of computers and more importantly development of more advanced machine learning algorithms and approaches such as deep learning,
new algorithms are developed to analyze and forecast time series data. The research question investigated in this article is that whether and how the newly developed deep learning-based algorithms for forecasting time series data, such as “Long Short-Term Memory (LSTM)”, are superior to the traditional algorithms. The empirical studies conducted and reported in this article show that deep learning-based algorithms such as LSTM outperform traditional-based algorithms such as
ARIMA model. More specifically, the average reduction in error rates obtained by LSTM was between 84 - 87 percent when compared to ARIMA indicating the superiority of LSTM to ARIMA. Furthermore, it was noticed that the number of
training times, known as “epoch” in deep learning, had no effect on the performance of the trained forecast model and it exhibited a truly random behavior.
more »
« less
- NSF-PAR ID:
- 10186768
- Date Published:
- Journal Name:
- 2018 17th IEEE International Conference on Machine Learning and Applications
- Page Range / eLocation ID:
- 1394 to 1401
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract. Machine learning is quickly becoming a commonly used technique for wind speed and power forecasting. Many machine learning methods utilize exogenous variables as input features, but there remains the question of which atmospheric variables are most beneficial for forecasting, especially in handling non-linearities that lead to forecasting error. This question is addressed via creation of a hybrid model that utilizes an autoregressive integrated moving-average (ARIMA) model to make an initial wind speed forecast followed by a random forest model that attempts to predict the ARIMA forecasting error using knowledge of exogenous atmospheric variables.Variables conveying information about atmospheric stability and turbulence as well as inertial forcing are found to be useful in dealing with non-linear error prediction. Streamwise wind speed, time of day, turbulence intensity, turbulent heat flux, vertical velocity, and wind direction are found to be particularly useful when used in unison for hourly and 3 h timescales. The prediction accuracy of the developed ARIMA–random forest hybrid model is compared to that of the persistence and bias-corrected ARIMA models. The ARIMA–random forest model is shown to improve upon the latter commonly employed modeling methods, reducing hourly forecasting error by up to 5 % below that of the bias-corrected ARIMA model and achieving an R2 value of 0.84 with true wind speed.more » « less
-
Network slicing will allow 5G network operators to o�er a diverse set of services over a shared physical infrastructure. We focus on supporting the operation of the Radio Access Network (RAN) slice broker, which maps slice requirements into allocation of Physical Resource Blocks (PRBs). We �rst develop a new metric, REVA, based on the number of PRBs available to a single Very Active bearer. REVA is independent of channel conditions and allows easy derivation of an individual wireless link’s throughput. In order for the slice broker to e�ciently utilize the RAN, there is a need for reliable and short term prediction of resource usage by a slice. To support such prediction, we construct an LTE testbed and develop custom additions to the scheduler. Using data collected from the testbed, we compute REVA and develop a realistic time series prediction model for REVA. Speci�cally, we present the X-LSTM prediction model, based upon Long Short-Term Memory (LSTM) neural networks. Evaluated with data collected in the testbed, X-LSTM outperforms Autoregressive Integrated Moving Average Model (ARIMA) and LSTM neural networks by up to 31%. X-LSTM also achieves over 91% accuracy in predicting REVA. By using X-LSTM to predict future usage, a slice broker is more adept to provision a slice and reduce over-provisioning and SLA violation costs by more than 10% in comparison to LSTM and ARIMA.more » « less
-
null (Ed.)Predicting workload behavior during execution is essential for dynamic resource optimization of processor systems. Early studies used simple prediction algorithms such as a history tables. More recently, researchers have applied advanced machine learning regression techniques. Workload prediction can be cast as a time series forecasting problem. Time series forecasting is an active research area with recent advances that have not been studied in the context of workload prediction. In this paper, we first perform a comparative study of representative time series forecasting techniques to predict the dynamic workload of applications running on a CPU. We adapt state-of-the-art matrix profile and dynamic linear models (DLMs) not previously applied to workload prediction and compare them against traditional SVM and LSTM models that have been popular for handling non-stationary data. We find that all time series forecasting models struggle to predict abrupt workload changes. These changes occur because workloads go through phases, where prior work has studied workload phase detection, classification and prediction. We propose a novel approach that combines time series forecasting with phase prediction. We process each phase as a separate time series and train one forecasting model per phase. At runtime, forecasts from phase-specific models are selected and combined based on the predicted phase behavior. We apply our approach to forecasting of SPEC workloads running on a state-of-the-art Intel machine. Our results show that an LSTM-based phase-aware predictor can forecast workload CPI with less than 8% mean absolute error while reducing CPI error by more than 12% on average compared to a non-phase-aware approach.more » « less
-
Few-shot machine learning attempts to predict outputs given only a very small number of training examples. The key idea behind most few-shot learning approaches is to pre-train the model with a large number of instances from a different but related class of data, classes for which a large number of instances are available for training. Few-shot learning has been most successfully demonstrated for classification problems using Siamese deep learning neural networks. Few-shot learning is less extensively applied to time-series forecasting. Few-shot forecasting is the task of predicting future values of a time-series even when only a small set of historic time-series is available. Few-shot forecasting has applications in domains where a long history of data is not available. This work describes deep neural network architectures for few-shot forecasting. All the architectures use a Siamese twin network approach to learn a difference function between pairs of time-series, rather than directly forecasting based on historical data as seen in traditional forecasting models. The networks are built using Long short-term memory units (LSTM). During forecasting, a model is able to forecast time-series types that were never seen in the training data by using the few available instances of the new time-series type as reference inputs. The proposed architectures are evaluated on Vehicular traffic data collected in California from the Caltrans Performance Measurement System (PeMS). The models were trained with traffic flow data collected at specific locations and then are evaluated by predicting traffic at different locations at different time horizons (0 to 12 hours). The Mean Absolute Error (MAE) was used as the evaluation metric and also as the loss function for training. The proposed architectures show lower prediction error than a baseline nearest neighbor forecast model. The prediction error increases at longer time horizons.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ICMLA.2018.00227
