< Key Hightlight >

As per AS-IS scenario, the global federated learning solutions market size to grow from USD 117 million in 2023 to USD 201 million by 2028, at a Compound Annual Growth Rate (CAGR) of 11.4% during the forecast period. Various factors such as the potential to enable companies to leverage a shared Machine Learning (ML) model collaboratively by keeping data on devices and the capability to enable predictive features on smart devices without impacting user experience and leaking private information are expected to offer growth opportunities for federated learning solutions during the forecast period.

To know about the assumptions considered for the study, download the pdf brochure

Market Dynamics

Driver: Ability to ensure better data privacy and security by training algorithms on decentralized devices

Federated learning is being researched by major companies and plays a critical role in supporting privacy-sensitive applications where the training data are distributed at the edge. Federated learning takes a step toward protecting users’ data by sharing model updates. Companies can no longer ignore the growing importance of data privacy and data security. The approach of federated learning has provided a new paradigm for applications leveraging data. Currently, data silos and the focus on data privacy are important challenges for AI, but federated learning could be a solution. It could establish a united model for multiple organizations while the local and sensitive data is protected so that they could benefit together without having to worry about data privacy. Federated learning has received a lot of attention in the way the technology tackles the challenge of protecting users’ privacy by decoupling of data provisioned at end-user equipment and Machine Learning (ML) model aggregation, such as network parameters of deep learning at a centralized server. With federated learning, privacy can be classified in two ways: global privacy and local privacy. Global privacy necessitates that the model updates generated at each round are private to all untrusted third parties other than the central server. At the same time, local privacy further requires that the updates are also private to the server.

Restraint: Lack of skilled technical expertise

The major issue confronting most organizations while incorporating ML in their business processes is the lack of skilled employees, including IT experts. Since federated learning is a new concept, it becomes difficult for employees to understand and implement federated learning models for training data. This is due to the lack of training provided to employees for implementing federated learning models. Recruiting and retaining technical resources have become a significant focus for several enterprises due to the lack of skilled people to develop and execute federated learning projects that involve complex techniques, such as ML. To develop more specialized skill sets and job descriptions as an industry. For example, organizations need engineers who can handle and understand the new federated learning architecture involved with deploying and maintaining ML models.

Opportunity: Capability to enable predictive features on smart devices without impacting user experience and leaking private information

Mobile phones, wearable devices, and autonomous vehicles are just a few of the modern distributed networks generating a wealth of data each day. Owing to the growing computational power of these devices—coupled with concerns related to transmitting private information—it is increasingly attractive to store data locally and push network computation to the edge devices. Federated learning is an emerging approach that helps companies easily collect and store data. Federated learning has the potential to enable predictive features on smartphones without diminishing the user experience or leaking private information. Edge devices, such as smartphones and IoT devices, can benefit from the on-device data without the data ever leaving the device, especially for computationally constrained devices where communication is a bottleneck with smaller devices. Today, industries, such as BFSI, healthcare and life sciences, and retail and eCommerce, collect gigantic amounts of data generated by consumer devices, including mobile phones, tablets, and personal laptops, on a daily basis. The federated learning approach provides a unique way to build such personalized models without intruding users’ privacy.

Challenge: Indirect information leakage

Privacy concerns serve to motivate the desire to keep raw data on each local device in a distributed Machine Learning (ML) setting. However, sharing other information such as model updates as part of the training process brings up another concern—the potential to leak sensitive user information. For instance, it is possible to extract sensitive text patterns, such as a credit card number, from a Recurrent Neural Network (RNN) trained on the user data. Unlike differential privacy protection, the data and the model itself are not transmitted, nor can they be guessed by the other party’s data. Hence, there is a little possibility of leakage at the raw data level. Federated learning exposes intermediate results, such as parameter updates from an optimization algorithm, such as Stochastic Gradient Descent (SGD). However, no security guarantee is provided, and the leakage of these gradients may actually reveal important information when exposed together with data structure, such as in the case of image pixels.

As per AS-IS scenario, among verticals, the manufacturing segment to grow at a the highest CAGR during the forecast period

The federated learning solutions market is segmented on verticals into BFSI, healthcare and life sciences, retail and eCommerce, energy and utilities, and manufacturing, and other verticals (telecommunications and IT, media and entertainment, and government). As per AS-IS scenario, the healthcare and life sciences vertical is expected to account for the largest market size during the forecast period. Moreover, the manufacturing vertical is expected to grow at the highest CAGR during the forecast period. With the increasing focus on Industrial Internet of Things (IIoT) and the rise in competition, manufacturing companies are prioritizing the analysis of data collected from numerous sources, including web, mobile, stores, and social media.

Europe to hold the largest market size during the forecast period

As per AS-IS scenario, Europe, followed by North America, is estimated to account for the largest market size in the federated learning solutions market during the forecast period respectively. Stringent data regulations and high focus on data privacy, focus on innovation through research, and rapid technology infrastructure advancements across verticals are the factors expected to drive the growth. These regions are early adopters of technologies and home to most of the existing federated learning solutions providers. The federated learning solutions market in APAC is projected to grow at the highest CAGR from 2023 to 2028. The increase in the adoption of emerging technologies, such as big data analytics, AI, and IoT, and ongoing developments to introduce data regulations, as well as focus on hyper-personalization and contextual recommendation in support of budding eCommerce markets in key countries such as China, India, and Japan are expected to drive the growth of federated learning solutions in the region.

Key Market Players

The federated learning solutions vendors have implemented various types of organic as well as inorganic growth strategies, such as new product launches, product upgradations, partnerships and agreements, business expansions, and mergers and acquisitions to strengthen their offerings in the market. The major vendors in the global federated learning solutions market include NVIDIA (US), Cloudera (US), IBM (US), Microsoft (US), Google (US), Owkin (US), Intellegens (UK), DataFleets (US), Edge Delta (US), Enveil (US), Lifebit (UK), Secure AI Labs (US), Sherpa.ai (Spain), Decentralized Machine Learning (Singapore), and Consilient (US).

Scope of the Report

This research report categorizes the federated learning solutions market based on application, vertical, and region.

By application:

• Drug Discovery
• Data privacy and Security Management
• Risk Management
• Shopping Experience Personalization
• Industrial Internet of Things (IIoT)
• Online Visual Object Detection
• Other Applications (video analytics, corporate IT, genomics, and anomaly detection)

By vertical:

• BFSI
• Healthcare and Life Sciences
• Retail and eCommerce
• Manufacturing
• Energy and Utilities
• Other Verticals (Telecommunications and IT, Media and Entertainment, and Government)

By region:

• North America
• Europe
• APAC
• RoW

Recent Developments:

• In March 2021, NVIDIA launched the NVIDIA AI Enterprise, a comprehensive software suite of enterprise-grade AI tools and frameworks optimized, certified, and supported by NVIDIA that run on VMware vSphere. NVIDIA AI Enterprise enables customers to reduce AI model development time from 80 weeks to just eight weeks and allows them to deploy and manage advanced AI applications on VMware vSphere.
• In February 2021, Enveil introduced new version of ZeroReveal 3.0. It delivers the homomorphic encryption-powered capabilities through an efficient and decentralized framework designed to reduce risk and address business challenges, including data sharing, collaboration, monetization, and regulatory compliance. The solution enhancements delivered in 3.0 releases strengthen integration, performance, and user experience features for both Enveil’s ZeroReveal Search and ZeroReveal Machine Learning product lines.
• In November 2020, NVIDIA Clara Train 3.1 introduces a flexible authorization framework that enhances security to ensure sensitive data is protected. It also includes a new administration tool that enables a 10x increase in algorithm experimentation to boost researcher productivity. Clara Train 3.1 new features help healthcare developers scale federated learning securely and boost research productivity.
• In October 2020, LiveRamp acquired DataFleets. The acquisition of DataFleets will enable LiveRamp customers to gain access to a powerful set of privacy-preserving technologies that can be configured based on business needs. With the acquisition, LiveRamp’s products and platforms will invest in exploring data collaboration solutions to keep consumer privacy at the forefront, accelerate planned innovations to the customer experience, and stay ahead of the competition.
• In May 2020, Owkin launched the COVID-19 Open AI Consortium (COAI). The consortium will enable advanced collaborative research and accelerate the clinical development of effective treatments for patients who are infected with COVID-19. In this project, Owkin used federated learning, aiming to help healthcare companies understand why drug efficacy varies from patient-to-patient, enhance the drug development process, and identify the best drug for the right patient at the right time, to improve treatment outcomes.
• In May 2020, IBM collaborated with NVIDIA to accelerate edge analytics and deploy applications at the edge. It will enable clients to unleash the power of accelerated AI computing at the edge with NVIDIA’s easy-to-deploy cloud-native software stack.
• In February 2020, A new consortium of the EU Innovative Medicines Initiative (IMI) was introduced to accelerate the development of AI in medicine. The project called BIGPICTURE will go on for a period of six years. Project participants will be leading European research centers, hospitals as well as major pharmaceutical industries. Owkin participated in the BIGPICTURE consortium through developing AI models to unveil signatures from Whole-Slide Images (WSIs). This will include model training toward predicting outcomes, such as genetic mutations, treatment response, recurrence, survival, and more. Owkin federated learning technologies will enable the collaborative development of AI models.
• In February 2020, Edge Delta partnered with Snowflake. The partnership introduced a new approach where Edge Delta's federated learning is connected with the infinite scalability of Snowflake’s cloud data platform. With this partnership, customers will no longer have to pick and choose subsets of data to enable real-time monitoring and security alerting. Instead, they can analyze that data 100X faster and improve privacy and security posture.
• In December 2019, NVIDIA partnered with Owkin and King’s College London. The aim of the partnership is to protect patient’s data through federated learning in the healthcare and life sciences sector. This partnership brings together the best players in life science and healthcare, ML, and data center infrastructure..
• In January 2019, Cloudera merged with Hortonworks. With this merger, Cloudera will deliver the first enterprise data cloud - unlocking the power of any data, running in any cloud from the Edge to AI, on a 100% open-source data platform. The merger will enable the Cloudera team to provide customers with a comprehensive solution-set to bring the right data analytics to data anywhere the enterprise needs to work, from the Edge to AI, Enterprise Data Cloud.