Bairong Inc. Wins Three Patents for Federated Learning and Voice Interruption

Bairong Inc, a leading independent AI-powered technology platform in China, was recently granted three national invention patents in China, including two for Federated Learning technologies and one for “voice interruption method and device.”

The two Federated Learning patents are “a prediction method and system based on an isolated forest training of Vertical Federation” and “a mobile device credit anti-fraud prediction method and system based on Federated Learning.” These cutting-edge technologies can provide improved data analysis capability for existing AI risk control and management of business operations, and increase the efficiency and accuracy of anomaly detection.

“With the gradual transparency and increasing adoption of privacy protection technology, there will be a rapid development in the next two years for interconnection, open-source technology and standard customization, which will enable financial institutions to apply more technologies in various application scenarios,” said Zhang Shaofeng, CEO of Bairong.

In its third quarter earnings announcement, Bairong indicated that it would invest more in privacy computing, Federated Learning and Convolutional Neural Networks, so as to further increase the number of its core customers and the revenue from each customer.

Recommended AI News: Binance and a Consortium Led by MDI Ventures Establish Joint Venture to Grow the Blockchain Ecosystem in Indonesia

Two Federated Learning patents to improve the credit evaluation model

For financial risk control, the goal of Federated Learning is to achieve joint modeling and improve the effect of the AI model, on the basis of ensuring data privacy, security, and compliance. However, in the process of anomaly detection, the existing technology makes it difficult for multi-party data to cooperate, resulting in limited means for anomaly detection.

The patent for “a prediction method and system based on an isolated forest training of Vertical Federation” can improve the efficiency and accuracy of anomaly detection through an unsupervised algorithm, enrich joint AI modeling with different financial institutions, and push anti-fraud ability to a new level.

For the second Federated Learning patent, Bairong has developed an anti-fraud technical solution between users’ mobile devices and financial institutions with privacy protection on both sides. Through the “mobile device credit anti-fraud prediction method and system based on Federated Learning”, Bairong can eliminate the risk of user privacy leakage from their mobile devices with the existing technology and provide users with more secure technical protection.

With the two patented technologies, Bairong can provide China’s leading financial institutions with upgraded digital transformation solutions, including the improved credit evaluation model, user’s rights adjustment plan with pre-calculating, and the optimized targeting and reaching strategy.

Recommended AI News: C Ventures Backed RTFKT to be Acquired by NIKE

“Voice interruption method and device” to provide a better user experience

With its innovative patent for “voice interruption method and device”, Bairong integrates model-customized Automatic Speech Recognition, Natural Language Understanding and Text-to-Speech into the softswitch system, which can significantly reduce the network data transmission loss, bring interruption and delay to millisecond level, provide millisecond-level voice interaction and feedback, and improve the user experience with smart voice interaction.

This patented technology has already been adopted by the “Baixiaorong” AI voice robot, which has been deployed by hundreds of financial institutions. Under various financial scenarios, the AI voice robot can provide superior customer service experience and support the digital transformation of financial services.

Recommended AI News: Brinc Closes $130 Million Funding, Led by Animoca Brands, to Launch Web 3.0-Focused Accelerators and Fuel Global Expansion

[To share your insights with us, please write to sghosh@martechseries.com]