eCommerce 2015 - 2015 Workshop: Machine Learning for eCommerce

The goal of the workshop is to discuss the state-of-the-art in applying machine learning to e-Commerce related domains. As machine learning matures, it is becoming increasingly front and center of both e-commerce and brick and mortar corporations. The workshop will bring together leaders from industry and academia to discuss the business and scientific challenges of using machine learning in commerce.
We are looking for contributions on the broadest definition of machine learning and e-commerce. We ideally look for both a solid machine-learning contribution to an important problem in (e-)commerce. We particularly encourage contributions that:
1) Address a specific aspect of importance in (e-)commerce. Examples include:
1-1) Understanding products from semi-structured product feeds
1-2) Understanding customers intent from unstructured, customer generated content (queries, reviews)
1-3) Control / optimization of business metrics via (e-)commerce experiences that match customers to products (search, recommendations, advertising)
1-4) Systematic improvement of systems and processes via experimentation
1-5) Attribution of customer-generated events / metrics (clicks, conversions, revenue) to specific components of a system (multi-touch attribution problem)
1-6) Provide guarantees on customer-facing / business metrics in a rapidly changing environment through automatic reconfiguration / retraining of learning systems
1-7) Enable tradeoffs between metrics operating a very different temporal scales: operational business metrics, and customer lifetime value
2) Address a theoretical contribution to an established field commonly used in (e-)commerce systems, or provide empirical evidence that a known problem can be solved in a novel way. A very partial and incomplete list of topics includes:
2-1) Sequential decision-making for lifetime value optimization
2-2) Churn and lifetime value predictions
2-3) Visitor Web behavior modeling and visualization
2-4) Intention and sentiment analysis
2-5) Recommendation and personalization systems
2-6) Multivariate and high cardinality anomaly detection in streaming data
2-7) Change detection in multivariate streaming data
2-8) Attribution modeling
2-9) Data cleansing - imbalanced data, categorical variables, missing values, dimensionality reduction
2-10) Visitor stitching
2-11) A/B and multiple hypothesis testing and experimentation
2-12) Off-policy evaluation and optimization
2-13) Bid optimization in online advertising
INVITED SPEAKERS
Ayman Farahat (Yahoo!)
Vivek Farias (MIT)
Nicolas Le Roux (Criteo)
Lihong Li (Microsoft Research)
Alessandro Magnani (-AT-WalmartLabs)
Muthu Muthukrishnan (Rutgers University)
Devavrat Shah (MIT)
SUBMISSION INSTRUCTIONS
We invite researchers from different subfields of machine learning (e.g., supervised & unsupervised learning, reinforcement learning, online learning, active learning), optimization, operations research, management sciences, and econometrics, as well as application-domain experts (from e.g., digital marketing, recommendation systems, online advertisement) to submit an extended abstract or a paper (between 4 to 8 pages in NIPS format) of their work to ecommerc...-AT-gmail.com. Accepted papers will be presented as posters or contributed oral presentations. Previously accepted work, including the NIPS-2015 accepted papers, are also welcomed, but please indicate where the paper was accepted or published.
ORGANIZERS
Esteban Arcaute (WalmartLabs)
Mohammad Ghavamzadeh (Adobe Research & INRIA)
Shie Mannor (Technion)
Georgios Theocharous (Adobe Research)