Throughout the week, there will be two hackathon tracks for those who have more advanced skills, and want to work in teams to produce something concrete at the end of the Indaba. We hope many people will choose to participate in the hackathon.
See details of two hackathon problems for you to choose from below, on classifying animals from the Serengeti or working with malaria data.
Hack session times (Labs: SCC 203/SCC 209)
- Sunday 25 August, 14:00-16:00 (with 16:00-18:30 optional)
- Monday, 26 August, 11:00-13:00,
- Tuesday, 27 August, 11:00-13:00, 16:30-18:30
- (Wednesday, 28 August, 14.00-18.30 optional)
- Thursday, 29 August, 14:30-18:30
- …and evenings, between the Indaba parties!
Track 1: Snapshot Serengeti
- Join the #serengeti channel on slack.
Can AI help wildlife conservation? What can you create that will help wildlife conservation? In the Serengeti Wildlife Conservation Hackathon, you will have access to thousands of geo-located, time-stamped and labelled images from more than 200 camera traps in the Serengeti. You’d be able to plot the migration of wildebeest, learn from the data which animals are nocturnal, and build computer vision models to spot zebras and many other species. But can you do more…?
This is also a talk related to this topic Thursday 29 August by Meredith Palmer and Stig Petersen, entitled “Computing, camera traps, and conservation: how can machine learning inform ecology?”
- See this project page to get a sense of the data, how it’s collected and the current crowdsourced labelling.
Co-ordinators and contact:
- Tejumade Afonja, Aya Salama, Ismael Kone, Stig Petersen, Jonathan Schwarz, Ulrich Paquet, Meredith Palmer
Track 2: IBM-Zindi Malaria Challenge
Website here! (Will go live on 25 August)
- Join the #malaria channel on slack.
DescriptionMalaria is thought to have had the greatest disease burden throughout human history, while it continues to pose a significant and disproportionate global health burden. Participants of this Challenge should apply machine learning tools, specifically reinforcement learning, to determine novel solutions which could impact malaria policy in Sub Saharan Africa.
Specifically, Challenge participants will submit Solutions via the Platform that will make determinations with respect to how combinations of interventions which control the transmission, prevalence and health outcomes of malaria infection, should be distributed in a simulated human population.
The Challenge consists of an in-person live challenge, hosted on the Zindi Platform (www.zindi.africa) during the Indaba Conference. Zindi will provide Authorised Participants with an explanation of the Challenge and sample code.
About IBM Research – Africa
Please visit https://www.research.ibm.com/labs/africa/ for more information about IBM Research Africa
Co-ordinator and contact:
- Amy Bray, Celina Lee, Oliver Bent, Sekou Remy