Adaptive Explainable Continual Learning Framework for Regression Problems with Focus on Power Forecasts

Abstract

Compared with traditional deep learning techniques, continual learning enables deep neural networks to learn continually and adaptively. Deep neural networks have to learn new tasks and overcome forgetting the knowledge obtained from the old tasks as the amount of data keeps increasing in applications. In this article, two continual learning scenarios will be proposed to describe the potential challenges in this context. Besides, based on our previous work regarding the CLeaR framework, which is short for continual learning for regression tasks, the work will be further developed to enable models to extend themselves and learn data successively. Research topics are related but not limited to developing continual deep learning algorithms, strategies for non-stationarity detection in data streams, explainable and visualizable artificial intelligence, etc. Moreover, the framework-and algorithm-related hyperparameters should be dynamically updated in applications. Forecasting experiments will be conducted based on power generation and consumption data collected from real-world applications. A series of comprehensive evaluation metrics and visualization tools can help analyze the experimental results. The proposed framework is expected to be generally applied to other constantly changing scenarios.

Continual Deep Learning for Time Series Modeling

“…The results show that CLOPS performs better in three scenarios than the two other frontier methods—GEM [ 134 ] and MIR [ 135 ]. He [ 111 ] also focused on the target-domain incremental application scenario and data-domain incremental application scenario of continual learning and described how their previous framework, CLeaR, can be extended to learn inputs successively. The framework utilizes the storage of buffered data by a novelty detector.…”

“…This survey will follow the taxonomy of Lange et al [17] and Mundt et al [103]. [93,94,[106][107][108][109][110][111][112][113][114][115][116] (note: if a paper uses two methods separately with similar satisfactory results, the paper will be listed under both groups).…”

“… Tree diagram of the taxonomy of the continual learning methods for sensor time series classification and forecasting [ 93 , 94 , 106 , 107 , 108 , 109 , 110 , 111 , 112 , 113 , 114 , 115 , 116 ] (note: if a paper uses two methods separately with similar satisfactory results, the paper will be listed under both groups). …”

“…He [111] presented the outline of their explainability module based on techniques like dimension reduction methods and visualization methods. Currently, there is not much research with a focus on the explainability of the continual learning algorithms [141], resulting in the problems like a lack of transparency for the CL modules.…”