# Pastprop-RNN: improved predictions of the future by correcting the past

arXiv: https://arxiv.org/abs/2106.13881 · HTML: https://mlsanity.com/paper/2106.13881/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 35 references that are in the map.
- First posted on arXiv 25 Jun 2021.

## Builds on (most cited first, 13 of 13)

- [Recurrent Neural Networks for Multivariate Time Series with Missing Values](https://mlsanity.com/paper/1606.01865/) (2016)
- [Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding](https://mlsanity.com/paper/1802.04431/) (2018)
- [LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection](https://mlsanity.com/paper/1607.00148/) (2016)
- [Doctor AI: Predicting Clinical Events via Recurrent Neural Networks](https://mlsanity.com/paper/1511.05942/) (2015)
- [Recurrent Neural Network Regularization](https://mlsanity.com/paper/1409.2329/) (2014)
- [Deep Learning is Robust to Massive Label Noise](https://mlsanity.com/paper/1705.10694/) (2017)
- [Understanding LSTM -- a tutorial into Long Short-Term Memory Recurrent Neural Networks](https://mlsanity.com/paper/1909.09586/) (2019)
- [Evaluating Real-time Anomaly Detection Algorithms - the Numenta Anomaly Benchmark](https://mlsanity.com/paper/1510.03336/) (2015)
- [TadGAN: Time Series Anomaly Detection Using Generative Adversarial Networks](https://mlsanity.com/paper/2009.07769/) (2020)
- [A Data Quality-Driven View of MLOps](https://mlsanity.com/paper/2102.07750/) (2021)
- [A Comparative Analysis of Forecasting Financial Time Series Using ARIMA, LSTM, and BiLSTM](https://mlsanity.com/paper/1911.09512/) (2019)
- [Machine Learning vs Statistical Methods for Time Series Forecasting: Size Matters](https://mlsanity.com/paper/1909.13316/) (2019)
- [Ensemble Method for Censored Demand Prediction](https://mlsanity.com/paper/1810.09166/) (2018)

## Built on by (most cited first, 1 of 1)

- [Kernel Corrector LSTM](https://mlsanity.com/paper/2404.18273/) (2024)

## Abstract (arXiv)

Forecasting accuracy is reliant on the quality of available past data. Data disruptions can adversely affect the quality of the generated model (e.g. unexpected events such as out-of-stock products when forecasting demand). We address this problem by pastcasting: predicting how data should have been in the past to explain the future better. We propose Pastprop-LSTM, a data-centric backpropagation algorithm that assigns part of the responsibility for errors to the training data and changes it accordingly. We test three variants of Pastprop-LSTM on forecasting competition datasets, M4 and M5, plus the Numenta Anomaly Benchmark. Empirical evaluation indicates that the proposed method can improve forecasting accuracy, especially when the prediction errors of standard LSTM are high. It also demonstrates the potential of the algorithm on datasets containing anomalies.