NIPS Workshop on Transactional Machine Learning and E-Commerce.

Exploring interactions between the Machine Learning, E-Commerce and Economics communities

NIPS 2014, Friday 12 December 2014, Montreal

Organisers: Amos Storkey, Jake Abernethy and Mark Reid


In the context of building a machine learning framework that scales, the current modus operandi is a monolithic, centralised model building approach. These large scale models have different components, which have to be designed and specified in order to fit in with the model as a whole. The result is a machine learning process that needs a grand designer. It is analogous to a planned economy.

There is an alternative. Instead of a centralised planner being in charge of each and every component in the model, we can design incentive mechanisms for independent component designers to build components that contribute to the overall model design. Once those incentive mechanisms are in place, the overall planner need no longer have control over each individual component. This is analogous to a market economy. The result is a transactional machine learning. The problem is transformed to one of setting up good incentive mechanisms that enable the large scale machine learning models to build themselves. It turns out that many of the issues in incentivised transactional machine learning are also common to the issues that turn up in modern e-commerce setting. These issues include issues of mechanism design, encouraging idealised behaviour while modelling for real behaviour, issues surrounding prediction markets, questions of improving market efficiencies, and handling arbitrage, issue on matching both human and machine market interfaces and much more. On the theoretical side, there is a direct relationship between scoring rules, market scoring rules, and exponential families via Bregman Divergences. On the practical side, the issues that turn up in auction design relate to issues regarding efficient probabilistic inference. The chances for each community to make big strides from understanding the developments in the others is significant. This workshop will bring together those involved in transactional and agent-based methods for machine learning, those involved in the development of methods and theory in e-commerce, those considering practical working algorithms for e-commerce or distributed machine learning and those working on financially incentivised crowd-sourcing. The workshop will explore issues around incentivisation, handling combinatorial markets, and developing distributed machine learning. However the primary benefit will be the interaction and informal discussion that will occur throughout the workshop.

Workshop Outline

The following is the workshop outline as it currently stands. This is subject to change at this stage.
  • 8.30 - 8:45 Introduction to Workshop (Jake Abernethy)
  • 8:45 - 9:10 Talk 1: David Parkes will look at Scoring rules and incentivization for crowdsourcing.
  • Spotlights
  • 9:15 - 9:25 Ping Jin, Russell Greiner, Muyu Wei and Gerald Haeubl -- Using Survival Prediction Techniques to Learn Consumer-Specific Reservation Price Distributions.
  • 9:25 - 9:35 Nihar Shah and Dengyong Zhou -- Double or Nothing: Multiplicative Incentive Mechanisms for Crowdsourcing
  • 9:35 - 9:45 Mehryar Mohri and Andres Munoz Medina: Revenue Optimization in Posted-Price Auctions with Strategic Buyers
  • 9:45 - 10:00 Mohammad T. Irfan and Luis E. Ortiz: Causal Strategic Inference in Networked Microfinance Economies; Hau Chan and Luis Ortiz Computing Nash Equilibria in Generalized Interdependent Security Games
  • 10:00 Coffee, Discussion and Posters.
  • 10:30 - 10:50 Talk 2: Satyen Kale. Who moderates the moderators.
  • 10:50 - 11:10 Talk 3: Aaron Roth will talk on Truthful, Private Convex Optimization.
  • 11:10 - 11:30 Talk 4: Bobby Kleinberg will discuss Incentivizing Exploration.
  • 11:30 Morning Discussion Period and 2 or 3 Interactive 2 minute Commentaries.
  • 12:00 End of Morning Session.

12:00 - 15:00 Lunch.

  • 15:00 - 15:15 Intro to the Afternoon: What is Transactional Machine Learning, and how does it relate to E-Commerce (Amos Storkey)
  • 15:15 - 16:00 Talk 5: Robin Hanson will talk on Log Market Scoring Rules and Prediction Markets
  • 16:00 - 16:30 Talk 6: David Wolpert will share his enthusiasm for Collectives
  • 16:30 Coffee, Discussion and Posters.
  • 17:00 - 17:20 Talk 7: Rafael Frongillo and Mark Reid. Risk Dynamics in Trade Networks.
  • 17:20 - 17:40 Talk 8: Sebastian Lahie will look at combinatorial market makers.
  • 17:40 - 18:00 Talk 9: Jenn Wortman Vaughan will talk on Automated Market Makers.
  • 18:00 Discussion Period and 2 or 3, 2 minute Commentaries.
  • 18:30 Summary, Conclusions and End of Workshop.

Additional posters. In addition to posters from the spotlight presentations, we have late breaking and other exciting posters from Ben Moran, Jinli Hu, Chris Berlind, Ruth Urne and Sindhu Kutty.


If you wish to contribute a late breaking poster, please submit it on the NIPSTrans2014 EasyChair site.


Amazon Europe Yahoo!Labs

We thank Yahoo! Labs and Amazon for sponsoring the workshop.