I just had a great time visiting Prof. Dr. Dean Eckles at MIT; I attended ACM-EC (a great conference), and Dean and I discussed a lot of ongoing work (causality, habits, indirect inference, and much more). Looking forward to more collaborations. Here is a picture of Boston taken from the MIT Sloan building:
Note that we have also been doing a lot of software development:
- We presented UltimateSim (simulations of the Ultimatum game) during ACM-EC; see: https://github.com/Nth-iteration-labs/ultimatesim
- We are currently working on a total overhaul of StreamingBandit, making it easier to use, more scalable, etc. For progress see: https://github.com/Nth-iteration-labs/streamingbandit
- Finally, we are wrapping up a package for the online (row-by-row) estimation of hierarchical models. Hope to finalize the package soon, but see progress here: https://github.com/L-Ippel/SEMA_extended
That’s it for now; we wish all of you a great summer!
We did a few updates of our site; primarily updating the projects section. We are now fully emerged with the JADS institute (www.jads.nl) as one of the research labs; so do expect more projects and updates soon!
But, that means you should feel free to come and visit us!
We have a number of cool updates. First of all, we are happy to share the posters we will be presenting shortly at http://www.sigecom.org/ec17/index.html:
- Poster on Bootstrap Thompson Sampling (Eckles & Kaptein) – check it out here.
- Poster on Ultimatum Game Simulations (Kaptein) – check it out here.
Next, we are proud to present that our work on comparing operative vs. meta-judgmental measures for prediction of customer responses to persuasive messages was accepted at the Journal of Consumer Marketing (JCM). It will be in print soon, but you can find an author copy here.
Finally, I am proud to introduce two new PhD students: Xynthia Kavelaars will be working with Joris Mulder and Maurits Kaptein on a PhD project regarding Bayesian Adaptive Clinical trials, and Ylva Hendriks will be working on a joint project with Sebastiaan Peek and Prof. Inge Bongers on personalizing eHealth applications.
There is another position open in our lab; have a look at: https://www.academictransfer.com/employer/UVT/vacancy/40654/lang/en/
We are happy to announce that we will be presenting two posters at http://www.sigecom.org/ec17/.
- First, a poster on Bootstrap Thompson Sampling; see https://arxiv.org/abs/1410.4009 (together with Dean Eckles).
- A poster on agent based simulations. See: https://github.com/Nth-iteration-labs/ultimatesim
Looking forward to meeting people at MIT!
We are piloting a new study; have a look at: https://survey.mnds.org/content/survey-picasso-2017
Any feedback is well-appreciated!
We are hiring! I am involved in a number of PhD projects for which we are still looking for good candidates. First, check out the vacancies for the “Data science & Humans” research track at JADS:
Next, at Tilburg University I am involved in two exciting projects. Frist, we are looking for applicants for a project on adaptive Bayesian clinical trials:
Second, we are looking for candidates for a great position focussed on investigating the effects of e-health interventions:
Please do apply if you are interested!
I am delighted to share the fact that today PLOS ONE will publish our new work on Lock-in feedback. Its called:
“Uncovering noisy social signals: using optimization methods from experimental physics to study social phenomena”
and its pretty cool! Here is an author copy of the work.
Also, note that we are hiring a number of PhD students for diverse projects (adaptive clinical trials, sequential decision making, human-computer bandit problems, etc.). If you are looking for a PhD position please do let us know (by sending a mail to maurits [at] maurits kaptein [dot] com).
I wanted to share with you the JADS seminar series that Linnet Taylor and I are organizing together. Every week, during lunch on Tuesday, we house a seminar at JADS. Here are a few of the talks: http://www.jads.nl/jads-seminars.html. The seminars are free to attend, and we have a great lunch so feel free to drop by if you see an interesting topic!
Also, we have just started teaching a new data mining course. You can find more information here.
Finally, we are still awaiting a number of reviews on recent submission; I will share these as soon as we know more.
For now, that’s it!
Robin van Emden, Davide Iannuzzi, and I just presented some of our work on lock-in feedback at Physics@Veldhoven. Here are a few pictures. First, the poster:
And second, our demo setup, with two “Heroes of Physics”:
For a more elaborate description of the setup check out: http://pavlov.tech/2017/01/21/flappy/