Are you a PhD student who is doing research that uses methods of causal inference (randomized trials, natural experiments, etc) on social media and user generated data?
If so, I have good news, of a last minute addition to the AAAI Spring Symp. on Observational Studies through Social Media and Other Human-Generated Content. This workshop, which takes place at Stanford on March 21-23, is a 3-day gathering that brings together researchers across disciplines to discuss approaches and issues for causal inference research.
The organizers have kindly offered space for a doctoral feedback session for PhD students planning to do related work. This informal gathering will bring together PhD students to get feedback on their research designs and work together to improve the quality of our methods. Since so many of the speakers and attendees are experienced at using causal inference methods, it’s the perfect context to share early stage work and get feedback on your research.
To illustrate the kind of people who are gathering: here are some of the invited speakers:
- Susan Athey, who wrote this paper on machine learning and causal inference for policy evaluation
- Jure Leskovec, who co-led this ACM workshop on causal inference
- Aron Culotta, co-author of this paper estimating the effects of exercise on mental health, from Twitter data
- Dean Eckles, co-creator of PlanOut, an open source system used to design and run experiments at Facebook
- Adam Glynn, co-author of this working paper on methods of causal inference with time-series cross-sectional data
How To Apply (Deadline Oct 19)
To apply, submit an extended abstract via the symposium website by Monday October 19th and indicate your interest in participating in the feedback session. Extended abstracts can be no longer than 4 pages. It’s probably a good plan to submit something with hopes of getting a poster, but I get the impression that it’s also fine to submit something more work in progress. In my case, I’m planning to submit a section of my PhD proposal where I outline a proposed study design, adjusted for AAAI format.
See you at Stanford in October!