Datami is motivated by the growth in mobile (and wired) demand for data and ISPs' (Internet Service Providers) increasing inability to meet this demand. To make it worse, the heavy usage concentrates on several peak hours in a day, forcing ISPs to overprovision according to that. Even charging by monthly usage overages, as AT&T and Verizon have started doing in 2011, will not mitigate that. Yet this fact can be exploited to help solve the problem of excessive demand.

Many types of mobile applications traffic are time-elastic. In our recent online survey, 40% say that they're willing to wait 5-10 minutes in streaming YouTube videos if they see 2/3rd cheaper monthly bills. If we were to provide the right price incentives to these users, they will defer their usage from peak hours to lower priced periods, thereby flattening out the demand curve and reducing the ISP's operational cost.

DataMi implements an end-to-end system to offer these time-dependent prices. Our system computes the optimal prices to offer, given estimates of user reactions to these prices. ISPs can then monitor the resulting usage pattern over the next day and adapt the next day's prices to any observed changes in user behavior. We ran a trial of our system with 50 Princeton users from April 2011 to February 2012. We acted as a resale ISP to the participants, paying their AT&T data bills and charging them according to the time-dependent prices offered. We also assessed user responses to a colored indicator on the device's home screen that indicated the current price. Our results show that:

1. users adjust their usage according to colored price indicators
2. optimized TDP can significantly reduce the peak-to-average ratio of daily usage patterns
3. due to discounts in off-peak periods, TDP can induce users to significantly increase their overall usage

The results of our trial were published in ACM SIGCOMM 2012. Details of the price calculation method were published in IEEE ICDCS 2011, while the system components were featured in demos at IEEE INFOCOM 2012 and ACM MobiSys 2012 A more general survey on Smart Data Pricing (SDP) appeared in IEEE Communications Magazine and will appear in ACM Computing Surveys (preprint). A full publication list and details on the DataMi project is available at http://scenic.princeton.edu/datami/.

Edge Artificial Intelligence and Machine Learning

• Secure Machine learning
• Edge AI Test Bed
• Hurts to Be Too Early: Benefits and Drawbacks of Communication in Multi-Agent Learning.
• Robotic Intervention Learning at the Edge

Fog Computing

• Clouds with Deadlines
• Completion-time-aware Scheduler
• Securing Tor
• Multi-resource fairness
• LTE Multicast
• Massive Machine-Type Communications
• Coded Computing
• Client Control of HetNets
• Shred and Spread
• Big Data Sharing
• Networked Drone Cameras

Smart Data Pricing

• Time-dependent Pricing
• Cloud Bidding
• Sponsored Data
• Cloud Virtual Service Provider
• Traded Data Plans
• Quota-aware Video Adaptation

Social Learning Networks

• Personalized Thread Recommendations for MOOC Forums
• Individualization
• Engagement Modeling
• Clickstream Analytics
• Social Learning Efficiency