Department of Computer Engineering

Batch learning approaches cannot cope with the concept drift inherent in the stock market stream data. On the other hand, incremental learning has not been fully explored as a solution to the problem of concept drift in the stock market stream data. We propose a method that detects changes in stock features and quickly adapts the model structure to mitigate the drawbacks of data shifts through incremental learning. Our model uses Gated Recurrent Unit (GRU) enabling the self-growth of layers and hidden units that can dynamically adjust to the changes in the data distribution. Self-growing model architectures are known to experience catastrophic forgetting. To counter this, we incorporate a control mechanism capable of activating or deactivating certain layers and units depending on the occurrence of concept drifts. In our methodology, the GRU model can also be substituted with other deep learning models, such as RNN and LSTM. Our method uses the difference between the PCA eigenvectors for the two consecutive data windows. Our approach offers a model that evolves dynamically according to the changes while ensuring memory and time efficiency through its incremental nature. We evaluated our methodology on the CSI 300 dataset in the open-source quantitative investment platform Qlib and compared it with other studies in the field.