human behavioral computing

In 5 years smartphones will disappear into wearables and  smartspaces. Computing will become more intelligent, more tuned to our needs -- it's going to get personal. The next wave in computing is anticipatory -- consider Google Now as a good starting point. We plan to study  emergent anticipatory systems. Ultimately, these systems will be embedded -- implantables will allow humans to freely interact for the first time. This is not a taught course. It's a mix of seminar and project. 

We will read some papers, come up with project ideas and build apps and write up a workshop style paper. The class is mostly project oriented.

prerequisite

The prerequisite for this class are CS65/165 smartphone programming and CS 74/174 machine learning.

grading

25% class presentation and participation
50% project (pitch ideas, proposal, design/implementation, demo)
25% write up workshop style paper: idea, design, implementation and evaluation

time

Time period: 2A  TuTh   2:00-3:50   x-period: W 4:15-5:05.
Lecturer: Andrew Campbell

week 1

Human behavioral computing, Andrew Campbell
Move over, Siri,  The Economist, Nov 30th 2013
From Smart to Cognitive Phones, IEEE Pervasive Computing, June 2012
A Survey of Mobile Phone Sensing, IEEE Communications Magazine, September, 2010.
Anticipatory Mobile Computing: A Survey of the State of the Art and Research Challenges, Veljko Pejovic, Mirco Musolesi , under review.

week 2

We will not meet on Tuesday.

Thursday: Protect pitches. Each student gets 10 minutes to pitch two ideas each. You can use slides for your presentation. Send me a two page write up -- one page per idea prior to class (10% of grade). Projects based on Google Glass or Android phones.

week 3

Tuesday: carry on with brainstorming

week 4

Tuesday: Chenren Xu, Sugang Li, Gang Liu, Yanyong Zhang, Emiliano Miluzzo, Yih-Farn Chen, Jun Li, Bernhard Firner. Crowd++: Unsupervised Speaker Count with Smartphones. To appear in Proc. of the ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2013), Sept. 9-12, 2013, Zurich, Switzerland.

Han, Seungyeop, Matthai Philipose, and Yun-Cheng Ju. NLify: lightweight spoken natural language interfaces via exhaustive paraphrasing. Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing. ACM, 2013.

Tuesday: carry on with brainstorming. Now we have two projects.

week 5

Tuesday: teams, what are we going to implement and the plan to demo.

Xuan Bao, Songchun Fan, Alexander Varshavsky, Kevin Li, and Romit Roy Choudhury. 2013. Your reactions suggest you liked the movie: automatic content rating via reaction sensing. In Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing (UbiComp '13). ACM, New York, NY, USA, 197-206

Thursday: Chenren Xu, Sugang Li, Gang Liu, Yanyong Zhang, Emiliano Miluzzo, Yih-Farn Chen, Jun Li, Bernhard Firner. Crowd++: Unsupervised Speaker Count with Smartphones. To appear in Proc. of the ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2013), Sept. 9-12, 2013, Zurich, Switzerland.

Adib, Fadel, et al. "3D Tracking via Body Radio Reflections." (2013).