Sequence Prediction Under Missing Data : An RNN Approach Without Imputation

TL;DR

This paper introduces a novel RNN-based method for sequence prediction that directly encodes missing data patterns without imputation, enabling effective forecasting even with incomplete data.

Contribution

It presents a lossless pattern encoding approach integrated into Encoder-Decoder RNNs for sequence forecasting with missing data, without relying on data imputation.

Findings

Effective handling of missing data in sequence prediction.

Versatile architecture applicable to classification and forecasting.

Validated on real datasets with natural and synthetic missingness.

Abstract

Missing data scenarios are very common in ML applications in general and time-series/sequence applications are no exceptions. This paper pertains to a novel Recurrent Neural Network (RNN) based solution for sequence prediction under missing data. Our method is distinct from all existing approaches. It tries to encode the missingness patterns in the data directly without trying to impute data either before or during model building. Our encoding is lossless and achieves compression. It can be employed for both sequence classification and forecasting. We focus on forecasting here in a general context of multi-step prediction in presence of possible exogenous inputs. In particular, we propose novel variants of Encoder-Decoder (Seq2Seq) RNNs for this. The encoder here adopts the above mentioned pattern encoding, while at the decoder which has a different structure, multiple variants are feasible. We demonstrate the utility of our proposed architecture via multiple experiments on both single and multiple sequence (real) data-sets. We consider both scenarios where (i)data is naturally missing and (ii)data is synthetically masked.