header background

Advanced Federated Learning for everyone

Leverage your business with the leading privacy preserving collaborative learning technology.

SHERPA.AI FEDERATED LEARNING
SOLUTION

Sherpa.ai Federated Learning Framework, with its proprietary B2B platform, is focused on extracting value from data with any source, preserving privacy, and ensuring compliance with current and future regulations.

protect the privacy of your users data with the sherpaai federated learning and differential privacy framework

Protect the privacy of your users' data with the Sherpa.ai Federated Learning and Differential Privacy Framework

With the Sherpa.ai Federated Learning and Differential Privacy Framework, sensitive data is processed on the client-side and never leaves it, ensuring privacy.

john sculley on sherpa.ais predictive analytics and privacy ai

John Sculley (former CEO of Apple) on Sherpa.ai’s Predictive Analytics and Privacy AI

Sherpa.ai is able to give the market assurance of the privacy of confidential data through its propietary Privacy-Preserving Federated Learning Framework and Platform, opening new opportunities to the AI world.

tom gruber says sherpaai is leading the way in privacy

Tom Gruber (co-founder & CTO of Siri) says Sherpa.ai is “Leading the Way” in Privacy

Not only does Sherpa.ai achieve highly personalized predictive intelligence services with its framework, but also manage to preserve the user privacy in the process.

USE
CASES

The Sherpa.ai Federated Learning and Differential Privacy Framework is an open-source framework for Machine Learning that allows collaborative learning to take place, without sharing private data. It has been developed to facilitate open research and experimentation in Federated Learning and Differential Privacy. Federated Learning is a Machine Learning paradigm aimed at learning models from decentralized data, such as data located on users’ smartphones, in hospitals, or banks, and ensuring data privacy.

This is achieved by training the model locally in each node (e.g., on each smartphone, at each hospital, or at each bank), sharing the model-updated local parameters (not the data) and securely aggregating them to build a better global model. Federated Learning can be combined with Differential Privacy to ensure a higher degree of privacy.

Differential Privacy is a statistical technique to provide data aggregations, while avoiding the leakage of individual data records. This technique ensures that malicious agents intervening in the communication of local parameters can not trace this information back to the data sources, adding an additional layer of data privacy.

This technology could be disruptive in cases where it is compulsory to ensure data privacy, as in the following examples:

  • When data contains sensitive information, such as email accounts, personalized recommendations, and health information, applications should employ data privacy mechanisms to learn from a population of users whilst the sensitive data remains on each user’s device.
  • When data is located in data silos, an automotive parts manufacturer, for example, may be reluctant to disclose their data, but would benefit from models that learn from other manufacturers' data, in order to improve production and supply chain management.
  • Due to data-privacy legislation, banks and telecom companies, for example, cannot share individual records, but would benefit from models that learn from data across several entities.

Sherpa.ai is focused on democratizing Federated Learning by providing methodologies, pipelines, and evaluation techniques specifically designed for Federated Learning. Hereunder there are some examples that illustrates the benefits of using our platform.

HORIZONTAL FEDERATED LEARNING

a physician with medical cap protector glasses and a face mask touching a hologram of dna

HEALTH

Lymphoma diagnosis


Hospitals are already using Sherpa.ai’s technology to obtain more accurate models, training it on the scans of several hospitals, with no data exchange.

This technology has proven to be successful in determining an accurate model for lymphoma detection.

a blue futuristic padlock with numbers and letters in the background

SECURITY

Learning with the use security cameras


Surveillance companies are already identifying people not wearing masks on various premises.

They have seen an increase in identification accuracy with Sherpa.ai’s Privacy-Preserving platform, which allows the AI model training, using the data sets of all cameras, without sharing any privacy-protected data or images.

ground view of skyscrapers made of crystal

BANKING

Fraud Detection


Banks with branches in different regions already take advantage of Sherpa.ai’s technology.

Through Sherpa.ai’s horizontal federated learning models, banks are able to mitigate risk and fraud and maximize revenue. This is possible thanks to the combination of each banks’ knowledge.

VERTICAL FEDERATED LEARNING​

a mobile phone and a laptop showing different investments

BANK AND INSURANCE

Subscription Forecast


Banks are benefitting from the knowledge of insurance companies, as they have customers in common. Prediction of long-term deposit subscriptions is being sharpened thanks to this extra knowledge.

Using Sherpa.ai’s Vertical Federated Learning models the knowledge of both companies is integrated without sharing any data so that this heterogeneity maximizes revenue

seven antennas in the dusk

INSURANCE AND TELCO

Purchase probability


Insurance companies are benefiting from Telco companies by increasing lead qualification and thus increasing revenue due to better conversion. With the Telco’s data about their own clients, companies can predict the probability of purchase due to someone’s browsing history for example.

Sherpa.ai’s Privacy-Preserving platform ensures the privacy of data and regulatory compliance while maintaining the highest accuracy rates.

FEDERATED TRANSFER LEARNING

three people sitting together pointing to the screen of a laptop

GENERAL

Numerical Example


Having different handwritten numbers and different quality mobile devices, resolution of the images varies across them.

With Sherpa.ai’s Federated Transfer Learning platform, transferring the knowledge from the good quality mobiles to other low-resolution phones and leveraging this knowledge is possible.