Processing of missing data by neural networks
This provides a general and theoretically justified solution for neural networks to process missing data without requiring complete training data, which is incremental but addresses a common bottleneck in data analysis.
The paper tackles the problem of handling missing data in neural networks by replacing the first hidden layer's neuron responses with their expected values, achieving better results than typical imputation strategies and other methods for incomplete data.
We propose a general, theoretically justified mechanism for processing missing data by neural networks. Our idea is to replace typical neuron's response in the first hidden layer by its expected value. This approach can be applied for various types of networks at minimal cost in their modification. Moreover, in contrast to recent approaches, it does not require complete data for training. Experimental results performed on different types of architectures show that our method gives better results than typical imputation strategies and other methods dedicated for incomplete data.