Summary of Large Scale Hierarchical Industrial Demand Time-series Forecasting Incorporating Sparsity, by Harshavardhan Kamarthi et al.
Large Scale Hierarchical Industrial Demand Time-Series Forecasting incorporating Sparsity
by Harshavardhan Kamarthi, Aditya B. Sasanur, Xinjie Tong, Xingyu Zhou, James Peters, Joe Czyzyk, B. Aditya Prakash
First submitted to arxiv on: 2 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Methodology (stat.ME)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This research paper proposes Hierarchical Adaptive Iterative Learning Scheduler (HAILS), a novel probabilistic hierarchical model that addresses two key challenges in demand forecasting applications at large companies. Firstly, it handles varying sparsity across the hierarchy by adaptively modeling sparse and dense time-series with different distributional assumptions. Secondly, it reconciles these models to adhere to hierarchical constraints, ensuring accurate and calibrated probabilistic forecasts. The proposed method is evaluated against real-world demand forecasting datasets, showing a significant 8.5% improvement in forecast accuracy and 23% better improvement for sparse time-series. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary HAILS is a new way of predicting what will happen in the future based on patterns we see in data from the past. This is important because many companies need to predict how much of certain products they will sell, and this helps them make good decisions about how to produce those products. The problem is that often these predictions are not very accurate, especially when there are many different types of products involved. HAILS solves this problem by taking into account the relationships between different products and how often they are bought. This makes the predictions much more accurate, which helps companies make better decisions. |
Keywords
* Artificial intelligence * Time series
