Oct 21, 2022

Applied Machine Learning Methods for Time Series Forecasting (AMLTS)

Workshop held in conjunction with CIKM 2022

Time series data is ubiquitous, and accurate time series forecasting is vital for many real-world application domains, including retail, healthcare, supply chain, climate science, e-commerce and economics. The choice of machine learning methods, both conventional and deep learning-based models, primarily depends on the nature of input data. In addition, several models have been adopted in industries with great success.

We invite quality, novel, and ingenious contributions within an industrial application setting. The papers may span across and are not limited to achievements addressing relevant forecasting challenges in retail, e-commerce, and online transaction systems. We invite submissions of long and short papers of two to eight pages (including references), representing actual industrial deployment, preliminary results, and proposals for new work in industry or academics. All submissions will be single-blind and peer-reviewed by an international program committee of researchers/industrial professionals with a high reputation. Accepted submissions will be required to be presented at the workshop.

Topics of interest on applied machine-learning time-series forecasting approaches include, but are not limited to:

Model/ Architecture-based:

- Effective classical and deep learning-based forecasting

- Probabilistic/Statistical forecasting models

- Novel/Enhanced approaches for short-term/long-term forecasting

- Change point detection/ extreme event forecasting models

- Outlier detection/removal in forecasting

- Bayesian models/ neural network models to quantify forecasting uncertainty.

Model Evaluation/Metrics:

- Quantifying the performance of the proposed method for forecasting output (causal inference, statistics methods.)

- Online/offline/ real-time based training/prediction models

- Existing/Improved evaluation metrics and performance study

Large-Scale Deployment:

- Scalable, automated forecasting pipeline applications.

- Best practices for sampling to solve scalability issues in forecasting

- Challenges and resolutions to scale forecasting models to big data.

Call for Papers

Paper Submission Deadline: August 15, 2022, 11:59 PM AoE.

Paper Notification: September 15, 2022, 11:59 PM AoE.

Camera Ready Version: September 30, 2022, 11:59 PM AoE.

Full-Day Workshop: October 21, 2022

This workshop follows the submission requirement by CIKM.

Instructions:

- Long paper (up to 8 pages) and short paper (up to 4 pages). The page limit includes the bibliography and any possible appendices.

- Single-blind peer review

- All papers must be formatted according to ACM sigconf template manuscript style, following the submission guidelines available at: https://www.acm.org/publications/proceedings-template.

- Papers should be submitted in PDF format, electronically, using the EasyChair submission system

- All selected papers will invited for presentation.

Inquiry Email: amlts.workshop@gmail.com

Keynote Speakers

Image

Recent Advances in Bayesian Structural Time Series

Steven Scott

Director of Data Science at ShareThis

Bayesian structural time series models are a powerful and flexible class of time series models. They combine additive models for trends, seasonality, regression effects, holidays, and potentially other components in an easy to use framework that allows advanced Bayesian Markov chain Monte Carlo techniques to account for issues like variable and model selection through spike and slab priors. By including additional latent variables the model can be extended to incorporate non-Gaussian error distributions such as student T, Poisson, and binomial observation errors, or student T errors in the state space. The original implementation of Bayesian structural time series was in the bsts R package, which continues to e widely used. Bsts has recently been ported to Python. The R and python packages are not perfect substitutes. The R package supplies many more state models for univariate time series modeling, while the python package contains more recent work involving multivariate series.

Speaker bio:
Steve Scott is the director of Data Science at Share This, a Palo Alto based startup focused on extracting information from a network of internet publishers. Dr Scott is a Bayesian statistician with a 25 year career spanning academics and industry. He received his PhD from the Harvard Statistics Department. He has served on the faculty of the USC Marshall School of Business, and has held director level positions at Capital One, Google, and multiple startups. Dr Scott's published work focuses on applied Bayesian computation, where he's made important contributions around hidden Markov models, multinomial logistic regression, support vector machines, sparse regression, time series modeling, and Bayesian approaches to reinforcement learning and multi-armed bandit problems.

Image

Forecasting Cyber Events

Paul Yoo

Deputy Director, Birkbeck Institute for Data Analytics, University of London, UK

Traditional methods to cyber defence have been reactive monitoring system's traffic or behaviours to find sequences and patterns that match a previously observed signature of attack, incident, threat and action. More recently, machine-learning-based anomaly-detection methods have gained popularity, however, the detection at best could be done in real-time only. Forecasting events related to a cyber-attack prior to or during the attack is an under-explored area in cyber defence. This talk introduces a novel machine-learning based proactive cyber defence strategy that leverages upon various big (literature) data and logs in order to forecast cyber events prior to any malicious activity occurs so that allow defence and security to better prepare for, anticipate and counter future cyber threats, thereby reducing the impact of an attack and its likelihood of success.

Speaker bio: Paul Yoo is currently the Deputy Director for Birkbeck Institute for Data Analytics (BIDA) within Birkbeck College which is part of the University of London, United Kingdom. Prior to this, he held academic/research posts in Cranfield (Defence Academy of the UK), Sydney (USyd) and South Korea (KAIST). In his career, he has amassed more than ninety prestigious journal and conference publications, has been awarded more than US$ 2.3 million in project funding, and a number of prestigious international and national awards for his work in advanced data analytics, machine learning and secure systems research, notably IEEE Outstanding Leadership Award, Rozetta Award, Emirates Foundation Research Award, and the ICT Fund Award. Most recently, he won the prestigious Samsung award for research to protect IoT devices using machine-learning approach and Research England’s Global Challenge Research Fund (GCRF). Paul currently serves as an Associate Editor for ACM Computing Surveys and IEEE Transactions on Sustainable Computing. He had served as an Editor for IEEE COMML (big data and machine learning areas) from 2014 to 2019. He is also a Founder and Chair of the BIDA's Threat Intelligence lab. This lab has set out on a journey to bridge the gap between the advancement of machine learning and the progression of cyber security with the objective of creating the next generation intelligent cyber defence and security research environment for future cyber warfare, including the applications of cyber physical systems (e.g., autonomous vehicle). The lab is now a home for 11 PhD students.

Image

On the retail floor - industrial ad inventory forecasting with local and global models

Konstantin Shmakov

Distinguished Scientist at Walmart

In Ad-Tech billions of ads are being shown to online users. For advertising systems forecasting these ad-opportunities and planning based on forecasts represent a central problem. We will review application of local and global forecasting models for forecasting ad-opportunities in retail space, challenges and opportunities that come with it, benefits of cross-learning with DNN and incorporating of external forecasting signal to improve predictability. We also review probabilistic forecasting, conformal predictions and assessing uncertainty and planning based on these forecasts. We will present an industry-scale study based on Walmart ad-platform and evaluation of forecasting models made in large-scale industrial settings.

Speaker bio:
Konstantin is a Distinguished Data Scientist at Walmart Labs. Previously he spend 10+ years at Yahoo Labs advertising sciences group focusing his research on forecasting, recommendations and reinforcement learning for large scale systems in digital advertising. Applications include large-scale forecasting systems, recommendation systems, programmatic buying and real time bidding. He received Ph.D. in Physics from University of Tennessee, Knoxville and worked in experimental particle physics at Stanford Linear Collider on new QED physics and particle physics experiments.

Image

Customized Transformers for Time Series Forecasting

Qingsong Wen

Leader at Alibaba DAMO Academy

Although Transformer-based methods have significantly improved state-of-the-art results for time series forecasting, they are not only computationally expensive but more importantly, are unable to capture the global view of time series (e.g., overall trend) and handle complicated periodical patterns (e.g., multiple periods, variable periods). To address these challenges, we design two customized Transformers for time series forecasting. The first is termed as Frequency Enhanced Decomposed Transformer (FEDformer, in ICML’22) by combining Transformer with the seasonal-trend decomposition technique, in which the decomposition method captures the global profile of time series while Transformers capture more detailed structures. Furthermore, we exploit sparse representation in Fourier transform in the FEDformer to enhance the performance as well as reduce complexity to a linear complexity. The second is termed as Quaternion Transformer (Quatformer, in KDD’22) with three major components: 1) learning-to-rotate attention (LRA) based on quaternions which introduces learnable period and phase information to depict intricate periodical patterns; 2) trend normalization to normalize the series representations in hidden layers of the model considering the slowly varying characteristic of trend; 3) decoupling LRA using global memory to achieve linear complexity without losing prediction accuracy. Compared with state-of-the-art methods, extensive experiments on multiple time series benchmark datasets show that FEDformer and Quatformer can significantly improve the forecasting performance.

Speaker bio:
Dr. Qingsong Wen is currently a Staff Engineer / Team Leader at DAMO Academy-Decision Intelligence Lab, Alibaba Group (U.S.), working in the areas of intelligent time series analysis, data-driven intelligence decisions, machine learning, and signal processing. Before that, he worked at Futurewei, Qualcomm, and Marvell in the areas of big data and signal processing, and received his M.S. and Ph.D. degrees in Electrical and Computer Engineering from Georgia Institute of Technology, Atlanta, USA. He has published over 30 top-ranked conference and journal papers, and won the First Place in 2022 ICASSP Grand Challenge (AIOps in Networks) Competition. He is an Associate Editor for Neurocomputing, Guest Editor for Pattern Recognition, Guest Editor for Applied Energy, and regularly served as an SPC/PC member of the major DM/ML/AI conferences including KDD, ICDM, AAAI, IJCAI, etc.

Image

Demand Forecasting in Retail

Arash Sangari

Senior Director at Walmart

Speaker bio:
Arash Sangari is Senior Director of Data Science in Walmart Global Tech. His team with ‪Prakhar Mehrotra‬ (VP of ML) is focused on research and development of consumer demand forecasting models for Walmart stores and e-comm channel to serve variety of business use-cases in the largest retailer in the world. He has extensive experiences in leveraging various machine learning approaches in building robust, accurate and scalable demand forecasting models for major retailers in US. He has +15 years of experience in applying machine learning methods in solving high impact problems in academia and industry, with a track record of 13 published patents and several journal papers in multi-disciplinary research areas.

Accepted Papers

- Baoyu Jing, Si Zhang, Yada Zhu, Bin Peng, Kaiyu Guan, Andrew Margenot and Hanghang Tong. Retrieval Based Time Series Forecasting

- Shah Muhammad Hamdi, Abu Fuad Ahmad and Soukaina Filali Boubrahimi. Multivariate Time Series-based Solar Flare Prediction by Functional Network Embedding and Sequence Modeling

- Vinayak Gupta and Srikanta Bedathur. Modeling Human Actions in Time-Stamped Activity Sequences

- Yuhang Wu, Mengting Gu, Lan Wang, Yu-san Lin, Fei Wang and Hao Yang. Event2Graph: Event-driven Bipartite Graph for Multivariate Time Series Forecasting and Anomaly Detection

- Yujing Wang, Pingping Lin, Zhuo Li, Congrui Huang, Bixiong Xu and Yunhai Tong. Multi-Task Classifier Sharing for Cross-Dataset Time-Series Classification

- Byeong Tak Lee, Joon-myoung Kwon and Yong-Yeon Jo. Can Knowledge Distillation Really Transfer Inductive Bias?

- Khaznah Alshammari, Shah Muhammad Hamdi, Ali Ahsan Muhummad Muzaheed and Soukaina Filali Boubrahimi. Forecasting Multivariate Time Series of the Magnetic Field Parameters of the Solar Events

- Alberto Matuozzo, Paul Yoo, Alessandro Provetti and Maria Kim. Machine Learning Methods for Equity Time Series Forecasting: A Compendium

- Kan Huang, Kai Zhang and Ming Liu. GreenEyes: An Air Quality Evaluating Model based on WaveNet

- Gloria Li, Simon Fong and Antonio J. Tallón-Ballesteros. A Novel Classification by Dual Regression Algorithm for Machine Learning over Time Series in Human Activity Recognition

- Kenneth Odoh. Real-time Anomaly Detection for Multivariate Data Streams

- Somayeh Zamani, Hamed Talebi, Gunnar Stevens. Time Series Anomaly Detection in Smart Homes: A Deep Learning Approach

WORKSHOP ORGANIZERS

Image

Linsey Pang

Salesforce
Image

Wei Liu

University of Technology Sydney
Image

LingFei Wu

Pinterest
Image

Kexin Xie

Salesforce
Image

Stephen Guo

Indeed
Image

Raghav Chalapathy

Walmart Global Tech
Image

Musen Wen

Walmart Global Tech

Program Committee

Philippe Laban, Salesforce

Zhen Zhao, AVEVA

Vinayak Gupta, IBM

Jianghong Zhou, Walmart Labs

Md Omar Faruk Rokon, University of California, Riverside

Dongxia Wu, University of California San Diego

Sunny Verma, University of Technology, Sydney

Yuxuan Liang, National University of Singapore

Lebo Wang, Pinterest

Yunce Zhao, Southern university of science and technology

Yanbing Xue, Walmart Labs