ARMDN: Associative and Recurrent Mixture Density Networks for eRetail Demand Forecasting

By Srayanta Mukherjee, Devashish Shankar, Atin Ghosh, Nilam Tathawadekar, Pramod Kompalli, Sunita Sarawagi, Krishnendu Chaudhury. 2018

33 citations recorded by the source corpus.

Accurate demand forecasts can help on-line retail organizations better plan their supply-chain processes. The challenge, however, is the large number of associative factors that result in large, non-stationary shifts in demand, which traditional time series and regression approaches fail to model. In this paper, we propose a Neural Network architecture called AR-MDN, that simultaneously models associative factors, time-series trends and the variance in the demand. We first identify several causal features and use a combination of feature embeddings, MLP and LSTM to represent them. We then model the output density as a learned mixture of Gaussian distributions. The AR-MDN can be trained end-to-end without the need for additional supervision. We experiment on a dataset of an year's worth of data over tens-of-thousands of products from Flipkart. The proposed architecture yields a significant improvement in forecasting accuracy when compared with existing alternatives.

Subfields: Management Science and Operations Research, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

In Signal Processing

Citations in the corpus

Deep Factors for Forecasting (2019, background): “Changing the emission probability to Gaussian Mixtures results in AR-MDN (Mukherjee et al., 2018).”
Forecasting Big Time Series: Old and New (2018, background): “Diﬀerent RNN forecasters are introduced [37, 6, 41, 17, 25] and we explain the intuitions behind diﬀerent structures and demonstrate their performances on a variety types of time series.”
Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (2020, methodology): “[136] or quantile functions Gasthaus et al.”
Probabilistic Forecasting with Spline Quantile Function RNNs (2019, methodology): “Our approach is closely related to (Flunkert et al., 2017; Mukherjee et al., 2018) in that we that we em-Probabilistic ploy an almost identical RNN architecture.”
Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting (2020, background): “Time-series forecasting is a critical ingredient across many domains, such as sensor network monitoring (Papadimitriou and Yu 2006), energy and smart grid management, economics and ﬁnance (Zhu and Shasha 2002), and disease propagation analysis (Matsubara et al. 2014).”