EconBase
← All papers

F-FOMAML: GNN-Enhanced Meta-Learning for Peak Period Demand Forecasting with Proxy Data

Zexing Xu, Linjun Zhang, Sitan Yang, Rasoul Etesami, Hanghang Tong, Huan Zhang, Jiawei Han

arXiv 23 Jun 2024 · Machine Learning · 2 citations (OpenAlex)

arXiv:2406.16221 · PDF · DOI · OpenAlex · Extracted main text

Abstract

Demand prediction is a crucial task for e-commerce and physical retail businesses, especially during high-stake sales events. However, the limited availability of historical data from these peak periods poses a significant challenge for traditional forecasting methods. In this paper, we propose a novel approach that leverages strategically chosen proxy data reflective of potential sales patterns from similar entities during non-peak periods, enriched by features learned from a graph neural networks (GNNs)-based forecasting model, to predict demand during peak events. We formulate the demand prediction as a meta-learning problem and develop the Feature-based First-Order Model-Agnostic Meta-Learning (F-FOMAML) algorithm that leverages proxy data from non-peak periods and GNN-generated relational metadata to learn feature-specific layer parameters, thereby adapting to demand forecasts for peak events. Theoretically, we show that by considering domain similarities through task-specific metadata, our model achieves improved generalization, where the excess risk decreases as the number of training tasks increases. Empirical evaluations on large-scale industrial datasets demonstrate the superiority of our approach. Compared to existing state-of-the-art models, our method demonstrates a notable improvement in demand prediction accuracy, reducing the Mean Absolute Error by 26.24% on an internal vending machine dataset and by 1.04% on the publicly accessible JD.com dataset.

Citation extraction

53
references
67
in-text mentions
53
distinct cited
3
self-citations
7,062
main-text words

appendix boundary found by appendix_command · 65% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Alex Nichol, Joshua Achiam, and John Schulman (2018) On First-Order Meta-Learning Algorithms0.8434475%
2Sitan Yang, Malcolm Wolff, Shankar Ramasubramanian, Vincent Quennevi… (2023) GEANN: Scalable graph augmentations for multi-horizon time series forecasting. In KDD 2023 Workshop on Deep Learning on Graphs self0.8434475%
3Chelsea Finn, Pieter Abbeel, and Sergey Levine (2017) Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks. In Proceedings of the 34th International Conference on Machin…0.64422100%
4Zhenguo Li, Fengwei Zhou, Fei Chen, and Hang Li (2017) Meta-sgd: Learning to learn quickly for few shot learning0.64422100%
5Risto Vuorio, Shao-Hua Sun, Hexiang Hu, and Joseph J. Lim (2019) Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation. In Neural Information Processing Systems0.64422100%
6Pan Zhou, Yingtian Zou, Xiaotong Yuan, Jiashi Feng, Caiming Xiong, a… (2020) Task Similarity Aware Meta Learning: Theory-inspired Improvement on MAML. In 4th Workshop on Meta-Learning at NeurIPS0.64422100%
7Boris N Oreshkin, Dmitri Carpov, Nicolas Chapados, and Yoshua Bengio (2020) N-BEATS: Neural basis expansion analysis for interpretable time series forecasting. In Proceedings of the 8th International Conf…0.58531100%
8Neo Wu, Bradley Green, Xue Ben, and Shawn O'Banion (2020) Deep transformer models for time series forecasting: The influenza prevalence case0.51121100%
9Huaxiu Yao, Ying Wei, Junzhou Huang, and Zhenhui Li (2019) Learning to learn by remembering. In Advances in Neural Information Processing Systems. 1574–15840.51121100%
10Ilya Sutskever, Oriol Vinyals, and Quoc V. Le (2014) Sequence to Sequence Learning with Neural Networks. In NIPS. 3104–31120.40511100%

Showing the top 10 of 53 scored citations.