Federated machine learning startup integrate.ai today announced the availability of its privacy-preserving machine learning and analytics platform.
The new integrate.ai platform users federated learning and differential privacy technologies to unlock a range of machine learning and analytics capabilities on data that would otherwise be difficult or impossible to access because of privacy, confidentiality or technical hurdles.
Integrate.ai argues that traditional machine learning and analytics approaches require centralization and aggregation of data sources, often necessitating data-sharing agreements and supporting infrastructure. Traditional methods can present a roadblock for data-driven problems in the healthcare, industrial and finance sectors, where the highest privacy and security standards are required to ensure regulatory and contractual compliance.
The new service is packaged as a tool that enables developers to seamlessly integrate its capabilities into almost any solution with an easy-to-use software development kit and supporting cloud service for end-to-end management. Once the service is integrated, end-users can collaborate across sensitive data sets while allowing data custodians to retain complete control. Solutions incorporating integrate.ai can serve as both effective experimentation tools and production-ready services, the company says.
Integrate.ai’s platform is designed to address the issue where collaboration barriers can be broken, since data doesn’t need to be moved. The platform allows data to stay distributed in its original protected environments while unlocking its value with privacy-protective machine learning and analytics. Operations such as model training and analytics are performed locally and only end results are aggregated in a secure and confidential manner, according to the company.
“When data can be securely accessed and collaborated upon, we unlock boundless opportunities for life-saving research and innovation,” Steve Irvine, founder and chief executive officer of integrate.ai, said in a statement. “By allowing organizations to work in a federated way, our platform helps reduce cost structure, accelerate progress against product roadmaps and capture new revenue opportunities — all with more speed and flexibility than any other solution on the market.”
In one use case, DNAstack Corp., a company that offers software for scientists to more efficiently find, access and analyze genomic and biomedical data, uses integrate.ai’s product platform to support federated learning in its work in autism. DNAstack is leading the Autism Sharing Initiative, a global collaboration push to create a federated network of autism data, empowering better genetic insights and accelerating precision healthcare approaches.
“Genetic and health datasets are large, sensitive and globally distributed, making it impossible to bring them all together in one place,” said Marc Fiume, co-founder and CEO of DNAstack. “Federated learning will empower us to ask new questions about autism across global networks while preserving privacy of research participants.”
Integrate.ai is a venture capital-backed company, having raised $49.6 million, according to Crunchbase. Investors include TELUS Ventures, Portage Ventures, Real Ventures and Georgian.