Musketeer project has just completed its first period. The first 18 months have delivered promising results with the first version of our industrial data platform. MUSKETEER mission is to develop a platform with scalable algorithms for federated and privacy-preserving machine learning techniques, detection and mitigation of adversarial attacks, and a rewarding model capable of fairly monetizing datasets according to the real data value. This work has shown first results in the manufacturing area with the establishment of a first use case involving Comau and FCA, with the support of Engineering, trying to improve the welding quality of robots without disclosing the data provided by multiple automotive factories. This achievement comes with first outcomes that can be shared with the community. IBM has just released a Musketeer Client Library enabling interactions with the IBM Musketeer cloud platform for federated machine learning. Good news this comes as an open source resource that everyone is invited to use. This release is the first from a series to come sketching the contours of a true privacy preserving platform for industrial data.
- For the complete story on our manufacturing use case click here
- IBM Musketeer client library
- To know more on privacy preserving technologies used in Musketeer
- Don’t forget to follow us on Twitter and LinkedIn to stay informed of our next releases and outcomes:
Twitter
Linkedin
